JP2021141629A - Control apparatus for ac rotary electric machine - Google Patents

Abstract

To provide a control apparatus for an AC rotary electric machine capable of preventing a real modulation rate from overshooting a target value of a modulation rate or a steady-state deviation from occurring with respect to the target value of the modulation rate.SOLUTION: The present invention relates to a control apparatus 1 for an AC rotary electric machine which calculates a real value Mr of a modulation rate which is a ratio of an amplitude of a fundamental wave component of an application voltage to a coil of a plurality of phases with respect to a half value of a power supply voltage VDC on the basis of a voltage command value and a detection value of the power supply voltage VDC, calculates a predictive value Mrp of the real value of the modulation rate on the basis of the real value Mr of the modulation rate, and changes a current command value in such a manner that the predictive value Mrp of the real value of the modulation rate becomes closer to a target value of the modulation rate.SELECTED DRAWING: Figure 2

Description

The present application relates to a control device for an AC rotary electric machine.

The control device of the AC rotary electric machine may control the amplitude of the voltage applied to the three-phase windings to an overmodulation state in which the amplitude exceeds half of the power supply voltage in order to improve efficiency and output. On the other hand, when controlled to the overmodulation state, the voltage applied to the windings includes a harmonic component, and the power supply current also contains a harmonic component. Further, in the power supply connection path connecting the inverter and the DC power supply, an LC resonance circuit is formed by the smoothing capacitor of the inverter, and when the frequency of the harmonic component of the power supply current matches the resonance frequency of the power supply connection path, the power supply current becomes The harmonic component is amplified, which may adversely affect the DC power supply and other devices connected to the DC power supply.

In the technique of Patent Document 1, the modulation factor is controlled to a target value in a region where the rotation speed of the motor is equal to or higher than the base rotation speed. Highly efficient driving is performed by setting the target value in the overmodulation region.

In the technique of Patent Document 2, the expected voltage utilization rate, which is the ratio of the required output voltage to the maximum applicable voltage, is estimated based on the rotation speed, the current command value, and the motor constant, and the expected voltage utilization rate is the voltage saturation limit. The current command value is corrected by field weakening control or the like so as not to exceed a predetermined value corresponding to.

Japanese Unexamined Patent Publication No. 2012-200073 Patent No. 5292995

However, in the technique of Patent Document 1, when the rotation speed of the motor increases from a region smaller than the base rotation speed and becomes larger than the base rotation speed, the switching temporarily delays the follow-up of the magnetic flux control and the modulation factor. Overshoot may occur. Due to the overshoot of the modulation factor, an excessively overmodulated state may occur, the harmonic component of the power supply current may increase, and the DC power supply and other devices connected to the DC power supply may be adversely affected.

Further, in the technique of Patent Document 2, since the expected voltage utilization rate is estimated using the motor constant, if there is an error in the motor constant, the expected voltage utilization rate becomes smaller or larger than the actual voltage utilization rate. If the expected voltage utilization rate is smaller than the actual voltage utilization rate, even if the current command value is corrected so that the expected voltage utilization rate does not exceed the predetermined value, the actual voltage utilization rate exceeds the predetermined value and becomes excessive. There is a possibility of overmodulation. If the expected voltage utilization rate is larger than the actual voltage utilization rate, the actual voltage utilization rate may fall below a predetermined value and the output may decrease. Further, since the expected voltage utilization rate is an estimated value of the current actual voltage utilization rate, the expected voltage utilization rate overshoots a predetermined value due to the tracking delay of the field weakening control as in Patent Document 1, and excessively. There is a possibility of overmodulation.

Therefore, the present application provides a control device for an AC rotating electric machine capable of suppressing that the actual modulation factor overshoots the target value of the modulation factor or causes a steady deviation with respect to the target value of the modulation factor. Aim.

The control device for an AC rotary electric machine according to the present application is a control device for an AC rotary electric machine that controls an AC rotary electric machine having a stator and a rotor provided with several-phase windings via an inverter.
A current detector that detects the current flowing through the multi-phase windings, and
A voltage detector that detects the power supply voltage supplied from the DC power supply to the inverter, and
The current command value calculation unit that sets the current command value, and
A voltage command value calculation unit that calculates a voltage command value based on the current command value and the current detection value, and a voltage command value calculation unit.
A switching control unit that turns on and off a plurality of switching elements of the inverter based on the voltage command value and applies a voltage to the windings of the plurality of phases.
Based on the voltage command value and the detected value of the power supply voltage, the actual value of the modulation factor, which is the ratio of the amplitude of the fundamental wave component of the applied voltage of the multi-phase windings to the half value of the power supply voltage, is calculated. Modulation rate calculation unit and
A predictive modulation factor calculation unit that calculates a predicted value of the actual value of the modulation factor based on the actual value of the modulation factor is provided.
The current command value calculation unit changes the current command value so that the predicted value of the actual value of the modulation factor approaches the target value of the modulation factor.

According to the control device of the AC rotary electric machine according to the present application, the actual value of the modulation factor is calculated based on the voltage command value and the detected value of the power supply voltage, and the actual value of the modulation factor is predicted based on the actual value of the modulation factor. Since the value is calculated, it is possible to suppress the occurrence of a steady deviation from the actual value of the modulation factor in the predicted value of the actual value of the modulation factor. Then, since the current command value is changed so that the predicted value of the actual value of the modulation rate approaches the target value of the modulation rate, it is possible to prevent the actual value of the modulation rate from causing a steady deviation from the target value of the modulation rate. It is possible to prevent an excessively overmodulated state. In addition, since the predicted value of the actual value of the modulation rate that predicts the behavior of the actual value of the modulation rate is used, it is possible to predict in advance that the actual value of the modulation rate will overshoot the target value of the modulation rate, and the target of the modulation rate. The amount of overshoot of the actual value of the modulation factor with respect to the value can be suppressed, and an excessive overmodulation state can be suppressed.

It is a schematic block diagram of the AC rotary electric machine and the control device of the AC rotary electric machine which concerns on Embodiment 1. FIG. It is a schematic block diagram of the control device of the AC rotary electric machine which concerns on Embodiment 1. FIG. It is a hardware block diagram of the control device of the AC rotary electric machine which concerns on Embodiment 1. FIG. It is a figure explaining the resonance circuit of the power-source connection path which concerns on Embodiment 1. FIG. It is a figure which shows the frequency characteristic of the power-source connection path which concerns on Embodiment 1. FIG. It is a block diagram of the current command value calculation part which concerns on Embodiment 1. FIG. It is a block diagram of the feedback controller of the current command value calculation unit which concerns on Embodiment 1. FIG. It is a block diagram of the prediction modulation rate calculation unit which concerns on Embodiment 1. FIG. It is a figure explaining the setting of the resonance operation region and the phase lead gain which concerns on Embodiment 1. FIG. It is a time chart explaining the control behavior which concerns on a comparative example. It is a time chart explaining the control behavior which concerns on Embodiment 1. FIG. It is a time chart explaining the adjustment of the phase lead gain which concerns on Embodiment 1. FIG.

1. 1. Embodiment 1
The control device 1 (hereinafter, simply referred to as the control device 1) of the AC rotary electric machine according to the first embodiment will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an AC rotary electric machine 2 and a control device 1 according to the present embodiment.

1-1. AC rotary electric machine The AC rotary electric machine 2 has a stator and a rotor provided with a plurality of phases of windings. In the present embodiment, three-phase windings Cu, Cv, and Cw of U phase, V phase, and W phase are provided. The three-phase windings Cu, Cv, and Cw are star-connected. The three-phase winding may be a delta connection. The AC rotary electric machine 2 is a permanent magnet type synchronous rotary electric machine, and the rotor is provided with a permanent magnet.

The AC rotary electric machine 2 includes a rotation sensor 16 that outputs an electric signal according to the rotation angle of the rotor. The rotation sensor 16 is a Hall element, an encoder, a resolver, or the like. The output signal of the rotation sensor 16 is input to the control device 1.

1-2. Inverter or the like The inverter 20 is a power converter that performs power conversion between the DC power supply 10 and the three-phase winding, and has a plurality of switching elements. In the inverter 20, the switching element 23H (upper arm) on the positive electrode side connected to the positive electrode side of the DC power supply 10 and the switching element 23L (lower arm) on the negative electrode side connected to the negative electrode side of the DC power supply 10 are connected in series. Three sets of series circuits (legs) are provided corresponding to the windings of each of the three phases. The inverter 20 includes three switching elements 23H on the positive electrode side and three switching elements 23L on the negative electrode side, for a total of six switching elements. A connection point in which the switching element 23H on the positive electrode side and the switching element 23L on the negative electrode side are connected in series is connected to the winding of the corresponding phase.

Specifically, in the series circuit of each phase, the collector terminal of the switching element 23H on the positive electrode side is connected to the electric wire 14 on the positive electrode side, and the emitter terminal of the switching element 23H on the positive electrode side is the collector of the switching element 23L on the negative electrode side. The emitter terminal of the switching element 23L on the negative electrode side is connected to the terminal and is connected to the electric wire 15 on the negative electrode side. The connection point between the switching element 23H on the positive electrode side and the switching element 23L on the negative electrode side is connected to the winding of the corresponding phase. As the switching element, an IGBT (Insulated Gate Bipolar Transistor) in which the diode 22 is connected in antiparallel connection, a MOSFET (Metal Oxide Semiconductor Field Effect Transistor) having the function of a diode in which the diode 22 is connected in antiparallel connection, or the like is used. The gate terminal of each switching element is connected to the control device 1. Each switching element is turned on or off by a control signal output from the control device 1.

The smoothing capacitor 12 is connected between the positive electrode side electric wire 14 and the negative electrode side electric wire 15. A power supply voltage sensor 13 for detecting the power supply voltage supplied from the DC power supply 10 to the inverter 20 is provided. The power supply voltage sensor 13 is connected between the positive electrode side electric wire 14 and the negative electrode side electric wire 15. The output signal of the power supply voltage sensor 13 is input to the control device 1.

The current sensor 17 outputs an electric signal corresponding to the current flowing through the windings of each phase. The current sensor 17 is provided on the electric wire of each phase connecting the series circuit of the switching element and the winding. The output signal of the current sensor 17 is input to the control device 1. The current sensor 17 may be provided in the series circuit of each phase.

A charge / dischargeable power storage device (for example, a lithium ion battery, a nickel hydrogen battery, an electric double layer capacitor) is used for the DC power supply 10. The DC power supply 10 may be provided with a DC-DC converter, which is a DC power converter that boosts or lowers the DC voltage.

1-3. Control device The control device 1 controls the AC rotary electric machine 2 via the inverter 20. As shown in FIG. 2, the control device 1 includes a rotation detection unit 32, a voltage detection unit 33, a current detection unit 34, an actual modulation rate calculation unit 35, a prediction modulation rate calculation unit 36, and a current command value calculation unit 37, which will be described later. It includes a voltage command value calculation unit 38, a switching control unit 39, and the like. Each function of the control device 1 is realized by a processing circuit provided in the control device 1. Specifically, as shown in FIG. 3, the control device 1 includes an arithmetic processing unit 90 (computer) such as a CPU (Central Processing Unit), a storage device 91 for exchanging data with the arithmetic processing unit 90, as a processing circuit. An input circuit 92 for inputting an external signal to the arithmetic processing unit 90, an output circuit 93 for outputting a signal from the arithmetic processing unit 90 to the outside, and the like are provided.

The arithmetic processing device 90 is provided with an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), various logic circuits, various signal processing circuits, and the like. You may. Further, as the arithmetic processing unit 90, a plurality of the same type or different types may be provided, and each processing may be shared and executed. The storage device 91 includes a RAM (Random Access Memory) configured to be able to read and write data from the arithmetic processing unit 90, a ROM (Read Only Memory) configured to be able to read data from the arithmetic processing unit 90, and the like. Has been done. The input circuit 92 is connected to various sensors and switches such as a power supply voltage sensor 13, a current sensor 17, and a rotation sensor 16, and an A / D converter or the like that inputs the output signals of these sensors and switches to the arithmetic processing device 90. I have. The output circuit 93 is provided with a drive circuit or the like to which an electric load such as a gate drive circuit for driving the switching element on and off is connected and a control signal is output from the arithmetic processing unit 90 to the electric load.

Then, in each function of the control units 32 to 39 of FIG. 2 included in the control device 1, the arithmetic processing unit 90 executes software (program) stored in the storage device 91 such as a ROM, and the storage device 91, It is realized by cooperating with other hardware of the control device 1 such as the input circuit 92 and the output circuit 93. The setting data such as the target value of the modulation factor used by each of the control units 32 to 39 and the like is stored in a storage device 91 such as a ROM as a part of software (program). Hereinafter, each function of the control device 1 will be described in detail.

<Rotation detection unit 32>
The rotation detection unit 32 detects the magnetic pole position θ (rotation angle θ of the rotor) and the rotation angular velocity ω of the rotor at the electric angle. In the present embodiment, the rotation detection unit 32 detects the magnetic pole position θ (rotation angle θ) and the rotation angular velocity ω of the rotor based on the output signal of the rotation sensor 16. In the present embodiment, the magnetic pole position is set in the direction of the north pole of the permanent magnet provided in the rotor. The rotation detection unit 32 is configured to estimate the rotation angle (magnetic pole position) based on the current information obtained by superimposing the harmonic component on the current command value without using the rotation sensor. It may be (so-called sensorless method).

<Voltage detector 33>
The voltage detection unit 33 detects the power supply voltage VDC supplied from the DC power supply 10 to the inverter 20. In the present embodiment, the voltage detection unit 33 detects the power supply voltage VDC based on the output signal of the power supply voltage sensor 13.

<Current detector 34>
The current detection unit 34 detects the currents Iur, Ivr, and Iwr flowing through the three-phase windings. In the present embodiment, the current detection unit 34 detects the currents Iur, Ivr, and Iwr flowing from the inverter 20 to the windings Cu, Cv, and Cw of each phase based on the output signal of the current sensor 17. Here, Iur is the U-phase current detection value, Ivr is the V-phase current detection value, and Iwr is the W-phase current detection value. The current sensor 17 may be configured to detect the two-phase winding current, and the remaining one-phase winding current may be calculated based on the detected value of the two-phase winding current. For example, the current sensor 17 may detect the V-phase and W-phase winding currents Ivr and Iwr, and the U-phase winding current Iur may be calculated by Iur = -Ivr-Iwr.

The current detection unit 34 converts the three-phase current detection values Iur, Ivr, and Iwr into the d-axis current detection values Idr and the q-axis current detection values Iqr on the d-axis and q-axis rotating coordinate systems. The rotation coordinate system of the d-axis and the q-axis is a two-axis rotation coordinate consisting of the d-axis defined in the direction of the detected magnetic pole position θ and the q-axis defined in the direction advanced by 90 ° in the electric angle from the d-axis. It rotates in synchronization with the rotation of the magnetic pole position of the rotor. Specifically, the current detection unit 34 performs three-phase two-phase conversion and rotational coordinate conversion on the three-phase current detection values Iur, Ivr, and Iwr based on the magnetic pole position θ, and performs d-axis current detection value Idr. And the current detection value of the q-axis is converted to Iqr.

<Current command value calculation unit 37>
The current command value calculation unit 37 calculates the current command value. In the present embodiment, the current command value calculation unit 37 calculates the current command value Ido on the d-axis and the current command value Iqo on the q-axis. The details of the processing of the current command value calculation unit 37 will be described later.

<Voltage command value calculation unit 38>
The voltage command value calculation unit 38 calculates the voltage command value based on the current command value and the current detection value. In the present embodiment, the voltage command value calculation unit 38 calculates the voltage command value Vdo on the d-axis and the voltage command value Vqo on the q-axis based on the current command value and the detected value of the current, and calculates the voltage command value Vdo on the d-axis and the q-axis. Based on the voltage command values Vdo and Vqo of, the three-phase voltage command values Vuo, Vvo and Vwo applied to the three-phase windings are calculated. In the present embodiment, the voltage command value calculation unit 38 includes a dq-axis voltage command value calculation unit 381, a voltage coordinate conversion unit 382, and a modulation unit 383.

The dq-axis voltage command value calculation unit 381 d-axis so that the d-axis current detection value Idr approaches the d-axis current command value Ido and the q-axis current detection value Iqr approaches the q-axis current command value Iqo. Current feedback control is performed to change the voltage command value Vdo of the above and the voltage command value Vqo of the q-axis by PI control or the like. In addition, feedforward control may be performed for non-interference between the d-axis current and the q-axis current.

The voltage coordinate conversion unit 382 performs fixed coordinate conversion and two-phase three-phase conversion on the voltage command values Vdo and Vqo of the dq axis based on the magnetic pole position θ, and the three-phase voltage command value Vuoc after the coordinate conversion. Convert to Vvoca and Vwoc. The three-phase voltage command values Vuoc, Vvoca, and Vwoc after the coordinate conversion become sine waves, and correspond to the three-phase voltage command value or the fundamental wave component of the applied voltage of the three-phase winding.

As shown in the following equation, the modulation factor M of the three-phase voltage command value after coordinate conversion is the amplitude VA of the three-phase voltage command value after coordinate conversion, which is the fundamental wave component, with respect to half the value of the power supply voltage VDC. The ratio. The modulation factor M is also the ratio of the amplitude VA of the fundamental wave component of the applied voltage of the three-phase winding or the voltage command value of the three-phase after modulation to the half value of the power supply voltage VDC.
M = VA × 2 / VDC ・ ・ ・ (1)

The modulation unit 383 applies amplitude reduction modulation to the three-phase voltage command values Vuoc, Vvoca, and Vwoc after the coordinate conversion of the sine wave, and calculates the final three-phase voltage command values Vuo, Vvo, and Vwo. .. The modulation unit 383 is a line of the three-phase voltage command value with respect to the three-phase voltage command value after the coordinate conversion when at least the modulation factor M of the three-phase voltage command value after the coordinate conversion is larger than 1. Amplitude reduction modulation is applied to reduce the amplitude of the three-phase voltage command value while maintaining the inter-voltage. It should be noted that the amplitude reduction modulation may not be performed.

As the amplitude reduction modulation method, various known methods such as third-order harmonic modulation, min-max method (pseudo-third harmonic modulation), two-phase modulation, and trapezoidal wave modulation are used. The third harmonic superimposition is a method of superimposing the third harmonic on the voltage command value of the three phases after the coordinate conversion. The min-max method is a method in which 1/2 of the intermediate voltage of the three-phase voltage command value after the coordinate conversion is superimposed on the three-phase voltage command value after the coordinate conversion. Two-phase modulation is a method in which the voltage command value of any one phase is fixed to 0 or the power supply voltage VDC, and the other two phases are changed so that the line voltage of the three-phase voltage command value after coordinate conversion does not change. Is.

<Overmodulation state>
In the present embodiment, the voltage command value calculation unit 38 includes a final three-phase including an overmodulated state in which the amplitudes of the final three-phase voltage command values Vuo, Vvo, and Vwo exceed half the value of the power supply voltage VDC. The voltage command values Vuo, Vvo, and Vwo of are calculated. When amplitude reduction modulation is performed, an overmodulation state occurs when the modulation factor M is larger than 2 / √3 (≈1.15). When the amplitude reduction modulation is not performed, the overmodulation state occurs when the modulation factor M is larger than 1. A state in which the amplitudes of the final three-phase voltage command values Vuo, Vvo, and Vwo are less than half the value of the power supply voltage VDC is referred to as a normal modulation state. The modulation factor M at the boundary between the normal modulation state and the overmodulation state is referred to as a boundary modulation rate (1.15 when amplitude reduction modulation is performed).

<Switching control unit 39>
The switching control unit 39 turns on and off a plurality of switching elements by PWM (Pulse Width Modulation) control based on the three-phase voltage command values Vuo, Vvo, and Vwo. The switching control unit 39 generates a switching signal for turning on / off the switching element of each phase by comparing each of the voltage command values of the three phases with the carrier wave. The carrier wave is a triangular wave that oscillates with an amplitude of a power supply voltage VDC / 2 centered on 0 at a carrier frequency. The switching control unit 39 turns on the switching signal when the voltage command value exceeds the carrier wave, and turns off the switching signal when the voltage command value falls below the carrier wave. The switching signal is transmitted to the switching element on the positive electrode side as it is, and the switching signal obtained by inverting the switching signal is transmitted to the switching element on the negative electrode side. Each switching signal is input to the gate terminal of each switching element of the inverter 20 via a gate drive circuit to turn each switching element on or off.

<Current command value calculation unit 37>
The current command value calculation unit 37 sets the current command value based on the target value Mo of the modulation factor. In the present embodiment, the target value Mo of the modulation factor is set to a constant value (for example, 1.21). The current command value calculation unit 37 calculates the basic value Ψob of the interlinkage magnetic flux command value by multiplying the target value Mo of the modulation factor by the power supply voltage VDC and dividing by the rotation angular velocity ω.

Specifically, as shown in FIG. 6 and the following equation, the current command value calculation unit 37 multiplies the target value Mo of the modulation factor by 1/2 × √ (3/2) and the power supply voltage VDC, and the rotation angular velocity ω. Divide by with to calculate the basic value Ψob of the interlinkage magnetic flux command value.
Ψob = Mo × 1/2 × √ (3/2) × VDC / ω ・ ・ ・ (2)

Then, as shown in FIG. 6 and the following equation, the current command value calculation unit 37 adds the interlinkage magnetic flux correction value Ψoc, which will be described later, to the basic value Ψob of the interlinkage magnetic flux command value, and adds the interlinkage magnetic flux command value Ψo. Is calculated.
Ψo = Ψob + Ψoc ・ ・ ・ (3)

The current command value calculation unit 37 calculates the current command value Ido on the d-axis and the current command value Iqo on the q-axis based on the interlinkage magnetic flux command value Ψo and the torque command value To. The current command value calculation unit 37 refers to the d-axis current setting data in which the relationship between the interlinkage magnetic flux command value Ψo and the torque command value To and the d-axis current command value Ido is set in advance, and the interlinkage magnetic flux calculated. The d-axis current command value Ido corresponding to the command value Ψo and the torque command value To is calculated. The current command value calculation unit 37 refers to the q-axis current setting data in which the relationship between the interlinkage magnetic flux command value Ψo and the torque command value To and the q-axis current command value Iqo is set in advance, and the interlinkage magnetic flux calculated. The current command value Iqo of the q-axis corresponding to the command value Ψo and the torque command value To is calculated.

The current command value calculation unit 37 performs feedback control for changing the current command value so that the predicted value Mrp of the actual value of the modulation rate, which will be described later, approaches the target value Mo of the modulation rate. In the present embodiment, when the predicted value Mrp of the actual value of the modulation factor exceeds the target value Mo of the modulation factor, the current command value calculation unit 37 weakens the torque output while maintaining the torque output of the torque command value To. When the current command value is changed in the direction of magnetic flux and the predicted value Mrp of the actual value of the modulation factor falls below the target value Mo of the modulation factor, the weakening magnetic flux is reduced while maintaining the torque output of the torque command value To. The current command value is changed in the direction of weakening. By feedback control, the degree of weakening magnetic flux can be adjusted, and the predicted value Mrp of the actual value of the modulation factor can be brought closer to the target value Mo of the variability rate while maintaining the torque output of the torque command value To.

As shown in FIG. 7 and the following equation, the current command value calculation unit 37 calculates the deviation ΔM of the predicted value Mrp of the actual value of the modulation factor with respect to the target value Mo of the modulation factor, and sets the deviation ΔM to 1/2 × √. (3/2) and the power supply voltage VDC are multiplied and divided by the rotational angular velocity ω to calculate the control value U. Then, the current command value calculation unit 37 integrates the value obtained by multiplying the control value U by the control gain Km with a conditional integrator, and calculates the integrated value as the interlinkage magnetic flux correction value Ψoc. The conditional integrator has a so-called anti-windup function. That is, when the interlinkage magnetic flux command value Ψo reaches the upper limit value (upper limit value of the operable width) of the interlinkage magnetic flux command value Ψo set in the d-axis current setting data, the integrator sets the integrated value. Hold without increasing, and integrate when the interlinkage magnetic flux command value Ψo reaches the lower limit value (lower limit value of the operable width) of the interlinkage magnetic flux command value Ψo set in the d-axis current setting data. Keep the value without decreasing it.
ΔM = Mo-Mrp
U = ΔM × 1/2 × √ (3/2) × VDC / ω ・ ・ ・ (4)
Ψoc = ∫ (Km × U)

The torque command value To may be calculated in the control device 1 or may be transmitted from an external device.

In the present embodiment, when the predicted value Mrp of the actual value of the modulation factor exceeds the target value Mo of the modulation factor, the current command value calculation unit 37 controls the current command values Ido and q on the d-axis by weakening the magnetic flux control. By adjusting the current command value Iqo of the shaft, the predicted value Mrp of the actual value of the modulation factor can be made to follow the target value Mo of the modulation factor. Further, when the predicted value Mrp of the actual value of the modulation factor is lower than the target value Mo of the modulation factor, the current command value calculation unit 37 controls to weaken the magnetic flux to weaken the predicted value of the actual value of the modulation factor. Mrp can be made to follow the target value Mo of the modulation factor. When the rotation angular velocity ω is smaller than the base rotation speed, the predicted value Mrp of the actual value of the modulation factor is smaller than the target value Mo of the modulation factor. However, since the operation range to the side where the weakening magnetic flux is weakened is limited, the predicted value Mrp of the actual value of the modulation factor remains below the target value Mo of the modulation factor.

<Actual modulation rate calculation unit 35>
The actual modulation rate calculation unit 35 calculates the actual value Mr of the modulation rate based on the voltage command value and the detected value of the power supply voltage VDC. In the present embodiment, the actual modulation factor calculation unit 35 uses the following equation to calculate the actual value Mr of the modulation factor based on the voltage command value Vdo on the d-axis, the voltage command value Vqo on the q-axis, and the power supply voltage VDC. calculate.

The actual modulation rate calculation unit 35 may calculate the actual value Mr of the modulation rate using the three-phase voltage command values Vuoc, Vvoca, and Vwoc after coordinate conversion as the voltage command value, and the final 3 The actual value Mr of the modulation factor may be calculated using the voltage command values Vuo, Vvo, and Vwo of the phase.

<Amplification of ripple component in overmodulation state due to resonance of power supply connection path>
In the overmodulated state where the amplitude of the above-mentioned three-phase voltage command value exceeds half the value of the power supply voltage VDC, the harmonic component superimposed on the line voltage of the applied voltage is the frequency of the fundamental wave (rotation frequency at the electrical angle). ) 5th and 7th order components become larger. On the other hand, as the harmonic component of the inverter current, the 5th and 7th order components of the applied voltage appear as the 6th order component.

In the overmodulation state, as the modulation factor M increases, the harmonic component superimposed on the line voltage of the applied voltage increases, and the torque ripple component and the harmonic component of the inverter current increase.

When the frequency of the sixth-order harmonic component of the inverter current generated in the overmodulated state matches the resonance frequency of the power supply connection path, the harmonic component of the power supply current is amplified and connected to the DC power supply 10 and the DC power supply 10. There is a risk of adversely affecting the device.

As shown in FIG. 4, the resonance circuit of the power supply connection path is an RLC series resonance circuit based on the capacitance C of the smoothing capacitor 12 of the inverter 20, the inductance L and the resistance R of the connection path between the DC power supply 10 and the smoothing capacitor 12. Is. The frequency characteristics are as shown in FIG. 5, and the gain is increasing in the resonance frequency band.

Therefore, in the overmodulation state, when the 6th order (6ω) frequency of the rotation angular velocity ω and the resonance frequency band of the power supply connection path overlap, the 6th order harmonic component of the power supply current is amplified. Further, even in the overmodulation state, as the modulation factor M increases, the amplitude of the 6th harmonic component before amplification increases, and in proportion to this, the amplitude of the 6th harmonic component after amplification also increases. growing. Therefore, in the resonance operation state, it is necessary to prevent the modulation factor M from becoming excessively larger than the boundary modulation factor between the normal modulation state and the overmodulation state.

In particular, in a transient state, when the actual value Mr of the modulation factor overshoots and exceeds the target value Mo of the modulation factor, the actual value Mr of the modulation factor becomes larger than the boundary modulation factor, and the sixth harmonic after amplification becomes larger. The amplitude of the wave component may increase. Therefore, it is desired that the actual value Mr of the modulation factor does not overshoot the target value Mo of the modulation factor too much.

<Predicted modulation rate calculation unit 36>
Therefore, the predicted modulation rate calculation unit 36 calculates the predicted value Mrp of the actual value of the modulation rate based on the actual value Mr of the modulation rate. Since the predicted value Mrp of the actual value of the modulation factor is calculated based on the actual value Mr of the modulation factor, it is possible to suppress the occurrence of a steady deviation of the predicted value Mrp of the actual value of the modulation factor with respect to the actual value Mr of the modulation factor. .. Then, as described above, the current command value is changed based on the predicted value Mrp of the actual value of the modulation factor. Therefore, since it can be predicted in advance that the actual value Mr of the modulation factor overshoots the target value Mo of the modulation factor by the predicted value Mr of the actual value of the modulation factor, the actual value of the modulation factor with respect to the target value Mo of the modulation factor can be predicted in advance. The amount of Mr overshoot can be suppressed. In particular, in the resonance operation region where the resonance of the current ripple component occurs in the power supply connection path, by suppressing the overshoot amount, the actual value Mr of the modulation factor is higher than the boundary modulation factor between the normal modulation state and the overmodulation state. Can be prevented from becoming too large, and an increase in the harmonic component of the power supply current can be suppressed.

In the present embodiment, as shown in the following equation, the predictive modulation factor calculation unit 36 performs phase lead processing on the actual value Mr of the modulation factor to calculate the predicted value Mrp of the actual value of the modulation factor. Here, Gad (s) is a transfer function of phase lead processing, ωh is a cutoff frequency, Kad is a phase lead gain, and s is a Laplace operator.

By this phase advance processing, the phase of the angular frequency ω in the range of the following equation is advanced. Therefore, as the phase advance gain Kad is increased, the range of the angular frequency ω at which the phase is advanced becomes wider on the low frequency side, and the effect of the phase advance becomes larger. Further, in the steady state, the transfer function Gad (s) of the phase lead processing becomes 1, so that the predicted value Mrp of the actual value of the modulation factor matches the actual value Mr of the modulation factor, and the actual value of the modulation factor. The predicted value Mrp does not have a steady-state deviation of the modulation factor from the actual value Mr.

In the present embodiment, as shown in FIG. 8 and the following equation, the predictive modulation factor calculation unit 36 performs high-pass filter processing and gain multiplication processing for multiplying the actual value Mr of the modulation factor by the phase lead gain Kad. This is performed to calculate the predicted addition value ΔMp, and the predicted addition value ΔMp is added to the actual value Mr of the modulation factor to calculate the predicted value Mrp of the actual value of the modulation rate. Here, Ghps (s) is a transfer function of high-pass filtering.

When the equation (8) is modified, it matches the equation (6). Therefore, the phase lead processing is performed by the high-pass filter processing and the gain multiplication processing.

In the present embodiment, in order to reduce a minute fluctuation component of high frequency included in the actual value Mr of the modulation factor, the predictive modulation factor calculation unit 36 performs high-pass filter processing and gain on the actual value Mr of the modulation factor. In addition to the multiplication process, a low-pass filter process is performed to calculate the predicted addition value ΔMp. Here, Glps (s) is a transfer function of low-pass filtering, and ωl is a cutoff frequency.

Since the low-pass filter processing performs the prediction based on the actual value Mr of the modulation factor in which the high-frequency fluctuation component is reduced, it is possible to suppress the deterioration of the prediction accuracy in response to the high-frequency fluctuation component.

The phase lead gain Kad, the cutoff angular frequency ωh of the phase lead processing and the high-pass filter processing, and the cut-off angular frequency ωl of the low-pass filter processing are determined by conformity. Since the modulation factor M changes according to changes in the power supply voltage VDC, the rotational angular velocity ω, and the torque command value To, the frequency band in which the phase is advanced is the power supply voltage VDC, the rotational angular velocity ω, and the torque command value To. It may be set to include the band of the rate of change of.

In the present embodiment, the predicted modulation factor calculation unit 36 limits the predicted value Mrp of the actual value of the modulation factor to the lower limit by the actual value Mr of the modulation factor. That is, as shown in the following equation, when the predicted value Mrp of the actual value of the modulation rate is less than the actual value Mr of the modulation rate, the prediction modulation rate calculation unit 36 modulates the predicted value Mrp of the actual value of the modulation rate. Set the actual value Mr of the rate.

When the high-pass filter processing and the gain multiplication processing are used, the predictive modulation factor calculation unit 36 limits the lower limit of the predicted addition value ΔMp to 0, and adds the predicted addition value ΔMp after the lower limit limit to the actual value Mr of the modulation factor. , Calculate the predicted value Mrp of the actual value of the modulation factor. That is, as shown in the following equation, the predictive modulation factor calculation unit 36 sets 0 in the predictive addition value ΔMp when the predictive addition value ΔMp is less than 0.

According to this configuration, the prediction is performed on the side where the actual value Mr of the modulation factor increases, and the prediction is not performed on the side where the actual value Mr of the modulation factor decreases. Therefore, it is possible to reduce the amount of overshoot on the increasing side where the actual value Mr of the modulation factor overshoots on the increasing side with respect to the target value Mo of the modulation rate. Therefore, in particular, the harmonic component increases, and the amount of overshoot to the overmodulation state side, which becomes a problem, can be reduced. On the other hand, although the amount of overshoot to the normal modulation state side cannot be reduced, there is no particular problem because the harmonic component is reduced on the safe side, and the influence of the prediction on the feedback characteristics can be suppressed.

Further, when the rotational angular velocity ω decreases, the actual value Mr also decreases. Therefore, when the lower limit is not limited, the predicted value Mrp of the actual value of the modulation factor becomes smaller than the target value Mo of the modulation factor, acts in the direction of increasing the actual value Mr of the modulation factor more than necessary, and weakens the magnetic flux control. May be inadequate. By the lower limit limiting process, it is possible to prevent the predicted value Mrp of the actual value of the modulation factor from becoming smaller than the target value Mo of the modulation factor, and it is possible to prevent the weakening magnetic flux control from becoming insufficient.

In the resonance operation region where resonance of the ripple component of the current occurs in the power supply connection path connecting the DC power supply 10 and the inverter 20, the predictive modulation factor calculation unit 36 has a phase in the phase lead processing in the phase advance processing as compared with the operation region other than the resonance operation region. Increase the degree of progress. According to this configuration, the overshoot suppression effect can be increased in the resonance operation region in which the harmonic component of the power supply current is amplified by resonance.

The predictive modulation factor calculation unit 36 increases the phase lead gain Kad in the gain multiplication process in the resonance operation region as compared with the operation region other than the resonance operation region.

The predictive modulation factor calculation unit 36 sets the phase lead gain Kad to a value larger than 0 in the resonance operation region, and sets the phase lead gain Kad to 0 in the operation region other than the resonance operation region to set the phase advance degree. Set to 0.

According to this configuration, in particular, in the resonance operation region where it is necessary to suppress the occurrence of overshoot, the phase lead processing is performed, and in the region other than the resonance operation region where the necessity is reduced, the phase advance degree is set to 0 and the phase advance is applied. The effect on the feedback characteristics can be eliminated.

In the present embodiment, as shown in FIG. 9, the resonance operation region is a region of the rotational angular velocity ω and the torque command value To in the overmodulated state, and the frequency of the sixth order (6ω) of the rotational angular velocity ω is It is set in a region close to the resonance frequency band of the power supply connection path and in which the sixth harmonic component of the power supply current after amplification becomes larger than a predetermined value.

<Control behavior>
FIG. 10 shows a comparative example in which the current command value is changed so that the actual value Mr of the modulation rate approaches the target value Mo of the modulation rate without using the predicted value Mr of the actual value of the modulation rate. Shows the control behavior of. When the torque command value To is constant, the rotational angular velocity ω (in the figure, the rotational angular velocity ω [rad / s] at the electric angle is converted to the rotational velocity [rpm] at the mechanical angle and displayed) is constant. It is rising with an inclination. The detected value of the rotational angular velocity ω is filtered to remove noise, and is smaller than the actual value of the rotational angular velocity ω. Therefore, the delay of the feedback system is large. The target value Mo of the modulation factor is set to 1.21 in the overmodulation state.

In the comparative example, since the predicted value is not used, the actual value Mr of the modulation factor overshoots to the increasing side with respect to the target value Mo. Since it is in an overmodulated state, an increase in the actual value Mr of the modulation factor increases the harmonic component superimposed on the line voltage of the applied voltage, and in particular, in the resonance operation region, the harmonic component of the power supply current after amplification increases. It becomes too large and may adversely affect the DC power supply 10 and the like.

Next, in FIG. 11, under the same conditions as in FIG. 10, the current command value is changed so that the predicted value Mrp of the actual value of the modulation factor approaches the target value Mo of the modulation factor. The control behavior of the form is shown. The predicted value Mrp of the actual value of the modulation factor overshoots the target value Mo of the modulation factor, but as a result, the overshoot of the actual value Mr of the modulation factor is suppressed.

FIG. 12 shows a change in the overshoot suppression effect due to a change in the phase lead gain Kad according to the present embodiment. Increasing the phase advance gain Kad increases the degree of phase advance and increases the effect of suppressing overshoot. On the other hand, when the phase lead gain Kad is reduced, the degree of phase lead is reduced and the overshoot suppression effect is reduced. Therefore, the phase lead gain Kad is adjusted so as to obtain the desired behavior.

<Example of diversion>
In the above embodiment, the case where the three-phase winding is provided has been described as an example. However, the number of phases of the winding may be set to an arbitrary number such as two phases and four phases as long as it is a plurality of phases.

In the above embodiment, a case where a set of three-phase windings and an inverter is provided has been described as an example. However, two or more sets of three-phase windings and inverters may be provided, and the same control as in each embodiment may be performed on each set of three-phase windings and inverters.

In the above embodiment, the case where the predicted modulation factor calculation unit 36 performs the phase lead processing on the actual value Mr of the modulation factor to calculate the predicted value Mrp of the actual value of the modulation factor has been described as an example. However, the predicted modulation rate calculation unit 36 may perform various prediction calculations on the actual value Mr of the modulation rate to calculate the predicted value Mr of the actual value of the modulation rate. For example, the predicted modulation rate calculation unit 36 may calculate the predicted value of the actual value of the modulation rate by adding the value obtained by multiplying the rate of change of the actual value Mr of the modulation rate by the gain to the actual value Mr of the modulation rate. good.

In the above embodiment, the case where the target value Mo of the modulation factor is set to a constant value has been described as an example. However, the current command value calculation unit 36 may change the target value Mo of the modulation factor according to the operating state. For example, the current command value calculation unit 36 refers to the target value setting data in which the relationship between the rotation angular velocity ω and the torque command value To and the target value Mo of the modulation factor is set in advance, and refers to the current rotation angular velocity ω and the torque command. The target value Mo of the modulation factor corresponding to the value To may be calculated.

In the above embodiment, the current command value calculation unit 37 uses the interlinkage magnetic flux command value as an intermediate parameter, changes the interlinkage magnetic flux command value based on the target value Mo of the modulation factor, and the like, and changes the interlinkage magnetic flux command value. The case where the current command value is set based on is described as an example. However, the current command value calculation unit 37 may set the current command value without using the interlinkage magnetic flux command value. For example, the current command value calculation unit 37 uses the voltage shortage ratio as an intermediate parameter and changes the voltage shortage ratio based on the target value Mo or the like of the modulation factor, as disclosed in Japanese Patent Application Laid-Open No. 2012-200073. Then, the current command value may be set based on the voltage shortage rate.

Although exemplary embodiments are described in the present application, the various features, aspects, and functions described in the embodiments are not limited to the application of a particular embodiment, either alone or. It can be applied to embodiments in various combinations. Therefore, innumerable variations not illustrated are envisioned within the scope of the techniques disclosed herein. For example, it is assumed that at least one component is modified, added or omitted.

1 Control device of AC rotary electric machine, 10 DC power supply, 32 rotation detection unit, 33 voltage detection unit, 34 current detection unit, 35 actual modulation rate calculation unit, 36 predictive modulation rate calculation unit, 37 current command value calculation unit, 38 voltage Command value calculation unit, 39 switching control unit, Kad phase lead gain, Mo modulation factor target value, Mr modulation factor actual value, Mrp modulation factor actual value prediction value, To torque command value, ΔMp prediction addition value, VDC Power supply voltage, ω rotation angle speed

Claims (11)

A control device for an AC rotary electric machine that controls an AC rotary electric machine having a stator and a rotor provided with multi-phase windings via an inverter.
A current detector that detects the current flowing through the multi-phase windings, and
A voltage detector that detects the power supply voltage supplied from the DC power supply to the inverter, and
The current command value calculation unit that sets the current command value, and
A voltage command value calculation unit that calculates a voltage command value based on the current command value and the current detection value, and a voltage command value calculation unit.
A switching control unit that turns on and off a plurality of switching elements of the inverter based on the voltage command value and applies a voltage to the windings of the plurality of phases.
Based on the voltage command value and the detected value of the power supply voltage, the actual value of the modulation factor, which is the ratio of the amplitude of the fundamental wave component of the applied voltage of the multi-phase windings to the half value of the power supply voltage, is calculated. Modulation rate calculation unit and
A predictive modulation factor calculation unit that calculates a predicted value of the actual value of the modulation factor based on the actual value of the modulation factor is provided.
The current command value calculation unit is a control device for an AC rotary electric machine that changes the current command value so that the predicted value of the actual value of the modulation factor approaches the target value of the modulation factor. The control device for an AC rotary electric machine according to claim 1, wherein the predicted modulation factor calculation unit performs phase lead processing on the actual value of the modulation factor to calculate the predicted value of the actual value of the modulation factor. The predictive modulation factor calculation unit calculates a predictive addition value by performing a high-pass filter process and a gain multiplication process of multiplying the gain on the actual value of the modulation factor, and the predictive addition is added to the actual value of the modulation factor. The control device for an AC rotary electric machine according to claim 1 or 2, wherein the values are added to calculate the predicted value of the actual value of the modulation factor. The control device for an AC rotary electric machine according to any one of claims 1 to 3, wherein the predicted modulation rate calculation unit limits the predicted value of the actual value of the modulation rate to the lower limit by the actual value of the modulation rate. The predictive modulation factor calculation unit calculates a predictive addition value by performing a high-pass filter process and a gain multiplication process for multiplying the gain on the actual value of the modulation factor, and limits the predictive addition value to a lower limit of 0. ,
The control of the AC rotary electric machine according to any one of claims 1 to 4, wherein the predicted addition value after the lower limit limit is added to the actual value of the modulation factor to calculate the predicted value of the actual value of the modulation factor. Device. According to claim 3 or 5, the predictive modulation factor calculation unit performs a low-pass filter process in addition to the high-pass filter process and the gain multiplication process on the actual value of the modulation factor to calculate the predictive addition value. The control device for the AC rotary electric machine described. In the resonance operation region where resonance of the ripple component of the current occurs in the power supply connection path connecting the DC power supply and the inverter, the predicted modulation factor calculation unit has a phase lead in the phase advance process more than other operation regions. The control device for an AC rotary electric machine according to claim 2, wherein the degree is increased. In the resonance operation region where resonance of the ripple component of the current occurs in the power supply connection path connecting the DC power supply and the inverter, the predictive modulation factor calculation unit obtains the gain in the gain multiplication process more than in other operation regions. The control device for an AC rotary electric machine according to any one of claims 3, 5, and 6 to be increased. The voltage command value calculation unit calculates the voltage command value of the multi-phase winding including the overmodulation state in which the amplitude of the voltage command value of the multi-phase winding exceeds half the value of the power supply voltage.
The control device for an AC rotary electric machine according to any one of claims 1 to 8, wherein the switching control unit turns on / off the plurality of switching elements based on voltage command values of the plurality of phases of windings. When the predicted value of the actual value of the modulation factor exceeds the target value of the modulation factor, the current command value calculation unit maintains the torque output of the torque command value and reduces the magnetic flux in the direction of the current. When the command value is changed and the predicted value of the actual value of the modulation factor falls below the target value of the modulation factor, the current command is directed in the direction of weakening the magnetic flux while maintaining the torque output of the torque command value. The control device for an AC rotary electric machine according to any one of claims 1 to 9, wherein the value is changed. When the predicted value of the actual value of the modulation factor exceeds the target value of the modulation factor, the current command value calculation unit reduces the interlinkage magnetic flux command value, and the predicted value of the actual value of the modulation factor becomes. If it is less than the target value of the modulation factor, the interlinkage magnetic flux command value is increased, and the current command value is calculated based on the interlinkage magnetic flux command value and the torque command value. The control device for the AC rotary electric machine described in item 1.

Images