JP6627702B2 - Control device for power converter - Google Patents

Description

  The present invention relates to a control device for a power converter.

  Conventionally, in a device that supplies power to a load such as an AC motor using a power converter such as an inverter, electric vibration occurs in an LC filter circuit arranged on the DC side of the power converter, and as a result, the voltage of the power converter also oscillates. It is known that the load control becomes unstable. Therefore, the control device for an AC motor disclosed in Patent Document 1 calculates the rate of change of the capacitor voltage, and operates the current command or torque command of the vector control unit by a damping operation or the like in accordance with the rate of change to obtain the voltage change of the capacitor. On the other hand, there is disclosed a method of controlling the current of the inverter to change in a direction to suppress the fluctuation.

Japanese Patent No. 4065901

  However, the method of Patent Document 1 has a problem that the compensator always operates in response to the voltage fluctuation of the capacitor, and the voltage fluctuation directly causes torque ripple. For example, when applied to a control device for an AC motor that requires a low torque ripple, such as an electric power steering of an automobile, the driver gives discomfort.

  The present invention has been made in view of such a point, and an object of the present invention is to provide a control device that can suppress torque ripple of an AC motor with respect to voltage fluctuation of a capacitor.

  The present invention relates to a control device (100, 200) for a power converter (44) having an inductance component (3) and a capacitor (4) on the DC power supply (2) side. The control device for the power converter includes an output voltage calculation unit (10, 11), a DC voltage detection unit (50), a current detection unit (51), a power calculation unit (20), and an output correction unit (23). , 24) and a reference voltage calculator (21).

The output voltage calculator calculates an output voltage command value (Vd ^* , Vq ^* ) to be output to the power converter.
The DC voltage detector detects a voltage across the capacitor, that is, a power converter input voltage (Vinv).
The current detector detects a current (Iinv) flowing through the power converter.
The power calculation unit calculates the input power (Pinv) of the power converter.
The output correction unit can correct at least one of the output voltage command value and the command value (ω ^* ) to be a control target in the calculation process of the output voltage calculation unit as a correction target command value.
The reference voltage calculator (21) calculates a reference voltage (Vref) from the circuit impedance (-r) from the DC power supply to the power converter and the input power of the power converter.

  When the input voltage of the power converter is equal to or lower than the reference voltage, the output correction unit corrects the correction target command value so that the current flowing through the power converter changes in a direction to suppress the fluctuation of the input voltage of the power converter. When the power converter input voltage is higher than the reference voltage, the correction target command value is maintained.

  When the input voltage of the power converter is equal to or lower than the reference voltage, the voltage fluctuation diverges and oscillates, and if the correction is not performed, the control of the motor becomes unstable. Conversely, when the power converter input voltage is higher than the reference voltage, the voltage fluctuation tends to stabilize.

Therefore, when the power converter input voltage is lower than the reference voltage, based on the difference between the power converter input voltage and the reference voltage, the current flowing through the power converter changes in a direction in which the fluctuation is suppressed, The output correction unit corrects a specified output voltage value or a command value to be a control target in a calculation process of the output voltage calculation unit.
On the other hand, when the input voltage of the power converter exceeds the reference voltage and the fluctuation of the voltage is stabilized, the output correction unit does not perform the correction.
Thereby, the torque ripple of the AC motor can be reduced as compared with the related art of Patent Literature 1 in which the fluctuation in the voltage applied to the power converter is always corrected.

It is a block diagram showing the composition of the power converter control device by a 1st embodiment. It is an equivalent circuit diagram from a DC power supply to a power converter. It is a flowchart of output correction control. It is a simulation result of the power consumption of the power converter control device in the prior art and the first embodiment. It is a block diagram showing the composition of the power converter control device by a 2nd embodiment. FIG. 3 is a frequency characteristic diagram of a closed loop transfer function of a current control system.

  Hereinafter, a plurality of embodiments in which a control device for a power converter of the present invention controls an AC motor will be described with reference to the drawings.

(1st Embodiment)
Control of an AC motor using the control device 100 of the power conversion controller of the present embodiment will be described. The inverter control device 100 as a “power converter control device” controls, for example, driving of an AC motor in a hybrid vehicle including an engine and an AC motor as power sources.
As shown in FIG. 1, an inverter 44 as a “power converter” is connected to the power supply main circuit 1.

  FIG. 2 shows a modeled power supply main circuit 1. The power supply main circuit 1 includes a DC power supply 2 and an LC composed of an inductance component 3 and a capacitor 4 for preventing the inflow of noise from the outside and for preventing the harmonic current due to the switching operation from flowing out to the power supply side. It has a filter circuit. The DC power supply 2 has a power supply internal resistance 5, and the inductance component 3 has a resistance component 7. As shown in the internal configuration diagram of the capacitor in FIG. 2, the capacitor 4 has a capacitance component 9 and an equivalent series resistance component (ESR) 6.

  The voltage across the capacitor 4 including the capacitance component 9 and the equivalent series resistance component (ESR) 6, that is, the voltage Vinv applied to the inverter is the inverter input voltage applied to the input of the inverter 44. The AC motor 55 is driven by the AC voltage output from the inverter 44. The control of the AC motor 55 is performed by a current feedback control method using vector control.

  Next, the vector control will be described. The vector control is to control the AC motor on a dq-axis rotating coordinate system in which an axis coinciding with the secondary magnetic flux axis of the AC motor is defined as a d-axis, and an axis orthogonal to the d-axis is defined as a q-axis. . First, parts other than the output correction unit 23, which are main parts of the present embodiment, will be described.

The output voltage calculator 10 receives, for example, a current command value I ^* calculated based on a torque command generated by a higher-level controller (not shown). I ^* is converted into a d-axis current command value Id ^* and a q-axis current command value Iq ^* by a current dq-axis conversion unit 25. The subtractor 26 subtracts the dq-axis current detection values Id, Iq fed back from the dq-axis current command values Id ^* , Iq ^* to calculate a dq-axis current deviation.

The current controller 27 generates dq-axis voltage command values Vd ^* and Vq ^* by PI control or the like so that the dq-axis current detection values Id and Iq follow the current command values Id ^* and Iq ^* , respectively.

The two-phase / three-phase converter 41 converts the output voltage command values Vd ^** and Vq ^** corrected by the output corrector 23 described later into three-phase voltage command values Vu ^* , Vv ^* and Vw ^* . The operation of the switching state of the inverter 44 based on the three-phase output voltage command values Vu ^* , Vv ^* , Vw ^* is performed by, for example, well-known PWM control.

When the inverter 44 applies desired three-phase AC voltages Vu, Vv, Vw to the AC motor by PWM control, the driving of the AC motor is controlled so that an output corresponding to the current command value I ^* is output.
The three-phase / two-phase converter 42 receives a phase current detection value from a current sensor 52 provided on a power line connected from the inverter 44 to the AC motor 55. In the present embodiment, the detected values of the V-phase current Iv and the W-phase current Iw are input from the current sensors provided for the V-phase and W-phase, and the remaining U-phase current Iu is estimated based on Kirchhoff's law. In other embodiments, a three-phase current may be detected, or a technique of estimating another two-phase current from one-phase current may be employed. The three-phase / two-phase converter 42 dq-converts the obtained three-phase current detection values Iu, Iv, Iw into dq-axis current detection values Id, Iq and feeds back to the subtracter 26. These are the driving methods of the electric motor by the current feedback control method using the general vector control.
Further, the inverter control device 100 includes a power calculation unit 20, a reference voltage calculation unit 21, and a subtractor 22. Their configuration and operation will be described in the following flowchart.

Next, the output correction unit 23 will be described. A method until the output correction unit 23 derives the correction amount will be described with reference to the output correction flowchart of FIG. Hereinafter, each step of deriving the correction amount is abbreviated as S.
In S101, the power calculation unit 20 acquires the inverter input voltage Vinv applied to the inverter 44 and the inverter current Iinv. The inverter input voltage Vinv is obtained from the DC voltage detector 50, and the inverter current Iinv is obtained from the current detector 51.
In S102, the power calculation unit 20 calculates the inverter input power Pinv by multiplying the inverter current Iinv by the inverter input voltage Vinv.

In S103, the reference voltage calculator 21 calculates the reference voltage Vref from the circuit impedance and the inverter input voltage Pinv. Details of the calculation method will be described later.
In S104, the subtractor 22 calculates a deviation between the reference voltage Vref and the inverter input voltage Vinv. The output correction unit 23 calculates a correction voltage amount by multiplying the deviation by a constant K. Details of the calculation method will be described later.
In S105, the adder 28 adds the correction voltage to the output voltage command values Vd ^* , Vq ^* to generate corrected output voltage command values Vd ^** , Vq ^** . Details of the correction method will be described later.
The outline of the output correction flowchart has been described above.

  Next, a detailed method for deriving the reference voltage Vref in S103 will be described. The equivalent circuit 1 from the DC power supply 2 to the inverter 44 is as shown in FIG. First, the relationship between the impedance of the equivalent circuit 1 and the power is obtained. Since the inverter 44 performs constant output control, the relationship among the inverter input voltage Vinv, the inverter current Iinv, and the inverter input power Pinv is expressed by Expression (1).

上式の両辺をＩｉｎｖで微分すると、式（２）が得られる。

When both sides of the above equation are differentiated by Iinv, equation (2) is obtained.

ここで、「−ｒ」は回路インピーダンスを表す。このように直流電源２の側から見てインバータ４４はかけた電圧に対して見かけ上負となる抵抗特性を持っている。なお、この負性抵抗ｒは周波数依存性を持たないものとして扱う。

Here, "-r" represents a circuit impedance. As described above, the inverter 44 has an apparently negative resistance characteristic with respect to the applied voltage when viewed from the DC power supply 2 side. The negative resistance r is treated as having no frequency dependency.

Next, the transfer function of the equivalent circuit 1 with the input being Vb (s) and the output being Vinv (s) is determined.
From Kirchhoff's current and voltage rules, the power supply voltage Vb (t) of the power supply 2 expressing the power balance of the equivalent circuit as a function of time t and the voltage Vc (t) applied only to the capacitance component 9 excluding the equivalent series resistance 6 And the circuit current Ib (t) are obtained as shown in Expressions (3), (4), and (5).

Ｒは電源内部抵抗５とリアクトル３の抵抗成分７を加えたものであり、Ｒ＝Ｒｂ＋Ｒｗで表される、Ｖｉｎｖ（ｔ）は時刻ｔの時にインバータ４４にかかる電圧、Ｌはリアクトル３のインダクタンスをそれぞれ表している。

R is the sum of the internal resistance 5 of the power supply and the resistance component 7 of the reactor 3, and is represented by R = Rb + Rw. Vinv (t) is the voltage applied to the inverter 44 at time t, and L is the inductance of the reactor 3. Each is represented.

Ｃはコンデンサ容量成分９を表している。Ｒｃはコンデンサの等価直列抵抗成分６である。

C represents a capacitor capacitance component 9. Rc is the equivalent series resistance component 6 of the capacitor.

式（３）、（４）、（５）をラプラス変換し、伝達関数Ｇ（ｓ）を求めると、以下の式（６）が求められる。
式（６）の導出は、例えば、非特許文献１に参照される通りである。
［非特許文献１］
直流給電システムの発振条件・発振領域に関する解析 田中 徹、山崎 幹夫
信学技法 ＴＥＣＨＮＩＣＡＬ ＲＥＰＯＲＴ ＯＦ ＩＥＩＣＥ ２００４−０７ Ｐ．２３〜２８

When the equations (3), (4), and (5) are subjected to Laplace transform and the transfer function G (s) is obtained, the following equation (6) is obtained.
The derivation of equation (6) is as described in Non-Patent Document 1, for example.
[Non-Patent Document 1]
Analysis on Oscillation Conditions and Oscillation Region of DC Power Supply System Toru Tanaka, Mikio Yamazaki IEICE Technical Report 2004-07 23-28

フルビッツの安定判別法より、式（６）の伝達関数Ｇ（ｓ）の伝達特性が安定であること、つまりＶｉｎｖの変動が安定するには式（６）括弧内の分母多項式係数のすべてが正でなければならない。

According to Hurwitz's stability discrimination method, in order for the transfer characteristic of the transfer function G (s) of equation (6) to be stable, that is, for the fluctuation of Vinv to be stable, all of the denominator polynomial coefficients in parentheses of equation (6) are positive Must.

式（６）の分母多項式全ての係数が正であるには、以下の式（８）または式（９）のいずれかで、それぞれ３つの不等式すべてを満足することが必要である。 At this time, the value of the estimated impedance Z is set as in Expression (7).

In order for the coefficients of all the denominator polynomials in Expression (6) to be positive, it is necessary that either of the following Expressions (8) or (9) satisfy all three inequalities.

  In order for the equation (9) to be satisfied, the circuit impedance “−r” needs to satisfy r> −Rb−Rw and r> −Rc. However, this does not hold when the law of maximum supply power is applied. Therefore, only when the equation (8) is satisfied, the fluctuation of the inverter input voltage Vinv is in a stable state.

  Here, from the equations (1) and (2), the reference voltage Vref can be obtained from the circuit impedance “−r” and the inverter input power Pinv as follows.

回路インピーダンスには式（８）を満足するＲｃ、Ｚ、Ｒのうちで最大の値を採用する。このとき、ＶｒｅｆはＶｉｎｖの変動が安定か、不安定かを判別するしきい電圧となる。Ｖｉｎｖ＜Ｖｒｅｆの場合、Ｖｉｎｖの変動は発振、発散する不安定領域にあり、Ｖｉｎｖ＞Ｖｒｅｆの場合、変動が収束するか、あるいは変動が拡大しない安定領域にある。
以上、基準電圧Ｖｒｅｆの算出に使用される回路インピーダンス「−ｒ」の算出方法について説明した。

As the circuit impedance, the maximum value among Rc, Z, and R satisfying Expression (8) is adopted. At this time, Vref is a threshold voltage for determining whether the fluctuation of Vinv is stable or unstable. When Vinv <Vref, the fluctuation of Vinv is in an unstable region where oscillation and divergence occur, and when Vinv> Vref, the fluctuation is in a stable region where the fluctuation converges or where the fluctuation does not expand.
The method of calculating the circuit impedance “−r” used for calculating the reference voltage Vref has been described above.

The derivation of the correction voltage in S104 will be described. When calculating the correction voltage, if the inverter input voltage Vinv is equal to or lower than the reference voltage Vref, the subtracter 22 subtracts the reference voltage Vref from the inverter input voltage Vinv to derive a deviation (Vinv-Vref), and then outputs the result. 23 multiplies the deviation by a constant K to generate a voltage correction value. The voltage correction values Vdc and Vqc are negative values, and the output voltage command values Vd ^** and Vq ^{** after} correction are smaller than Vd ^* and Vq ^* before correction.
When the inverter input voltage Vinv is larger than the reference voltage Vref, the voltage fluctuation is in the stable region, and the output correction unit 23 does not perform the output correction, so that the output voltage command values Vd ^* and Vq ^* are maintained. You.

Next, the correction method in S105 will be described.
As shown in the equation (2), “dVinv / dIinv” in the constant output operation indicates a negative resistance. Therefore, when the inverter input voltage Vinv decreases, the inverter current Iinv increases. Here, how the inverter input voltage Vinv is stabilized by reducing the inverter current Iinv will be described. Since there are two cases, each will be described.

(A) When both the inverter input voltage Vinv and the inverter current Iinv decrease The inverter current Iinv decreases as the inverter input voltage Vinv decreases. Therefore, “dVinv / dIinv”, that is, the r value becomes positive, and the negative resistance characteristic becomes Disappears. Therefore, all the terms on the right side in the equation (6) are positive. All the coefficients of the characteristic equation have the same sign, and the stability condition is always satisfied, so that the fluctuation of the inverter input voltage Vinv tends to be stable.

(B) When the inverter input voltage Vinv decreases and the inverter current Iinv increases When the increase rate of the inverter current Iinv is relatively smaller than the decrease rate of the inverter input voltage Vinv, "dVinv / dIinv" , That is, the absolute value of r becomes large, and the stability condition of Expression (8) is easily satisfied.
In either of the correction amounts (a) and (b), the increase in the inverter current Iinv is smaller than before the correction, and thus the fluctuation of the inverter input voltage Vinv tends to be more stable.

  FIG. 4 shows simulation results of the input power fluctuation of the inverter when this embodiment is performed and when it is not performed. The reference voltage Vref is a threshold voltage for determining whether or not the voltage fluctuation applied to the inverter is stabilized. FIG. 4A shows a case where Vinv fluctuates up and down around the reference voltage Vref. In the conventional method shown in FIG. 4B, the output correction is always performed for the voltage fluctuation applied to the inverter. As a result, the correction is performed even in a region where the power supply system becomes stable, and the inverter power (≒ AC motor torque) is reduced as a whole. It is getting bigger. On the other hand, when the present embodiment is implemented as shown in FIG. 4C, the output correction is not performed in the region where the power supply system is stable, but is corrected only in the region of the hatch portion where the power system becomes unstable. Therefore, the torque ripple can be reduced by half as compared with the conventional method.

  As described above, even if the fluctuation of Vinv becomes unstable, it is possible to stably control the output of the inverter by performing the output correction. Since the output correction is performed only when the inverter input voltage Vinv fluctuates in a direction lower than the reference voltage Vref, the torque ripple can be reduced as compared with the related art in which the fluctuation of the inverter input voltage Vinv is always corrected.

The output voltage command values Vd ^** and Vq ^** corrected by the output correction unit 23 are smaller than the output voltage command values Vd ^* and Vq ^* before correction. Therefore, actual current values Id and Iq flowing through the AC motor are smaller than current command values Id ^* and Iq ^* . Here, if the actual current values Id, Iq are made to follow the current command values Id ^* , Iq ^* by the current feedback control, the output voltage command values Vd ^* , Vq ^* are increased by the current controller 27 so as to increase. Will be corrected. When Vd ^* and Vq ^* are corrected to increase, the inverter current Iinv increases. Therefore, the value of “dVinv / dIinv” becomes smaller, and the absolute value of r also becomes smaller, which makes it difficult to satisfy Expression (8), and the fluctuation of Vinv cannot be stabilized.

  Therefore, the output correction unit 23 preferably sets the value of the constant K and the like so that the current controller 27 creates a negative voltage correction value larger than the correction value for increasing the voltage by the current feedback control. As a result, the inverter current Iinv can be effectively reduced, so that the fluctuation of the inverter input voltage Vinv can be prevented from becoming unstable.

Generally, the LC resonance frequency of the inverter input voltage Vinv fluctuation is often higher than the controllable frequency of the current control system. At this time, if the correction is performed on the output voltage command values Vd ^* and Vq ^* as in the present embodiment, there is no frequency dependency and the correction is easy.

Here, attention is paid to the second one of the three inequalities that need to be established in order to stabilize the fluctuation of Vinv from Expression (8). The second equation is expressed as follows.

  When the power supply voltage Vb is high, the capacitor capacitance component C of the equivalent circuit 1 generally increases. Therefore, the absolute value of the first term on the right side of Expression (11) becomes small. Therefore, the circuit impedance “−r” for satisfying the expression (11) may be small. In other words, the fluctuation of the inverter input voltage Vinv is more easily stabilized.

  Conversely, if the power supply voltage Vb is low, the fluctuation of the inverter input voltage Vinv tends to be unstable. Therefore, the present control device is more effective when employed in a power supply system having a small power supply voltage. For example, it is preferable to employ this control device when the DC power supply has a low voltage of 100 V or less in an automobile driven by an inverter from a DC power supply.

(2nd Embodiment)
FIG. 5 illustrates a control method of the AC motor 55 using the control device 200 of the power conversion controller according to the second embodiment. The operation of each unit will be described. In the second embodiment, substantially the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted.

The output voltage calculation unit 11 receives, for example, an angular velocity command value ω ^* generated by a higher-level control unit (not shown).

  The position sensor 53 is, for example, a resolver, and acquires the motor position θ of the AC motor 55. The differentiator 54 differentiates the motor position θ and converts it into an actual angular velocity value ω.

The speed command correction unit 31 calculates a new angular speed command value ω ^** from the angular speed correction amount ωc calculated by the output correction unit 24 described later, the angular speed command value ω ^*, and the actual angular speed value ω. The speed controller 32 generates a torque command value T ^* from ω ^** .

Based on T ^* , the torque current converter 33 generates a current command value I ^* with reference to a torque current map or the like.

When the inverter 44 applies the three-phase AC voltages Vu, Vv, Vw generated by the PWM control to the AC motor 55, the drive of the AC motor 55 is controlled so that an output corresponding to the angular velocity command value ω ^* is output. The operation of the output correction unit 24 will be described.
The operation up to the generation of the reference voltage Vref shown up to S103 in FIG. 3 is the same as that of the first embodiment. Thereafter, the reference voltage Vref is compared with the inverter input voltage Vinv. If Vref is large, the output correction unit 24 changes from the deviation in a direction in which the fluctuation is suppressed with respect to the fluctuation of Vinv, that is, in a direction in which Iinv decreases. Thus, the angular velocity correction amount ωc is generated, and the velocity instruction correction unit 31 corrects the angular velocity command value ω ^* .

As described above, in the second embodiment, the output voltage command value Vd ^*, Vq ^* itself rather than the output voltage command value Vd ^*, the angular velocity command value ω is a command value as a control target in the calculation process of Vq ^{* * The} output is corrected for ^* .

At this time, the input of the current control system of the output voltage calculation unit 11 is dq-axis current command values Id ^* and Iq ^* , and the output is the dq-axis actual current values Id and Iq, and the transfer function of the closed loop is obtained by the following equation.

この式の周波数特性は、図６のような一次遅れの応答になる。たとえばインバータ入力電圧Ｖｉｎｖ変動のＬＣ共振周波数が低い場合、つまり図６の１に相当する周波数領域では、実電流値Ｉｄ、Ｉｑは電流指令値Ｉｄ^*、Ｉｑ^*に追随し、有効な出力補正ができる。一方、図６の２に相当するＬＣ共振周波数が高い領域では、１の領域に対し、電流制御系の周波数特性のゲインが低下し、位相が遅れている。つまり、実電流値Ｉｄ、Ｉｑが電流指令値Ｉｄ^*、Ｉｑ^*に追随しないことがわかる。

The frequency characteristic of this equation has a first-order lag response as shown in FIG. For example, when the LC resonance frequency of the inverter input voltage Vinv fluctuation is low, that is, in the frequency region corresponding to 1 in FIG. 6, the actual current values Id and Iq follow the current command values Id ^* and Iq ^* , and effective output correction is performed. it can. On the other hand, in the region where the LC resonance frequency corresponding to 2 in FIG. 6 is high, the gain of the frequency characteristic of the current control system decreases and the phase is delayed with respect to the region 1. That is, it can be seen that the actual current values Id and Iq do not follow the current command values Id ^* and Iq ^* .

Therefore, if the LC resonance frequency of the inverter input voltage Vinv fluctuation is high, the correction to the angular velocity command value ω ^* cannot effectively control the inverter input voltage Vinv fluctuation. For example, in the case of a motor for driving an automobile, the LC resonance frequency is several kHz to several tens kHz, but it is difficult to accurately perform the response of the current control system at that frequency.

  Even in such a case, the correction can be performed by measuring the gain and phase for each frequency in advance and deriving the correction amount from the map data.

(Other embodiments)
(A) In the present embodiment, only the method of performing the output correction on the output voltage command values Vd ^* , Vq ^* , and the angular speed command value ω ^* is disclosed. However, the torque command value T ^* , the current command values Id ^* , Iq It is also possible to perform output correction on ^* .

  (B) In the present embodiment, the power converter 44 has been described as an inverter that converts DC power into AC power. However, the power converter is not limited to an inverter, and may be a DC converter that converts DC power into DC power. Is also good.

(C) The present invention is applicable to a system including a boost converter between a DC power supply and a capacitor, and also applicable to a system in which a capacitor voltage is supplied to another power source such as a DC converter or an air conditioner. It is. In addition, although only one power converter is used in the present embodiment, the present invention can be applied to a system in which a plurality of power converters are connected in parallel. Further, the present invention can be applied to loads other than the AC motor.
As described above, the present invention is not limited to the above embodiment, and can be implemented in various forms without departing from the spirit of the invention.

3 Reactor (Inductance Component) 4 Capacitors 10, 11 Output Voltage Calculator 20 Power Calculator 21 Reference Voltage Calculators 23, 24 Output Corrector 44 Inverter (Power Converter)
50 DC voltage detector 51 Current detector 100, 200 Inverter controller (power converter controller)

Claims (4)

A control device (100, 200) for a power converter (44) having an inductance component (3) and a capacitor (4) on a DC power supply (2) side,
An output voltage calculator (10, 11) for calculating an output voltage command value (Vd ^* , Vq ^* ) to be output to the power converter;
A DC voltage detector (50) for detecting a power converter input voltage (Vinv), which is a voltage across the capacitor;
A current detector (51) for detecting a current (Iinv) flowing through the power converter;
A power calculator (20) for calculating an input power (Pinv) of the power converter;
An output correction unit (23, 24) capable of correcting at least one of the output voltage command value and a command value (ω ^* ) to be a control target in a calculation process of the output voltage calculation unit as a correction target command value;
A reference voltage calculator (21) for calculating a reference voltage (Vref) from a circuit impedance (-r) from the DC power supply to the power converter and an input power of the power converter;
With
The output correction unit,
When the power converter input voltage fluctuates below the reference voltage, the correction target command value is corrected such that the current flowing through the power converter changes in a direction to suppress the fluctuation of the power converter input voltage,
A control device for a power converter that maintains the correction target command value when the power converter input voltage fluctuates more than the reference voltage.   The control device for a power converter according to claim 1, wherein the output correction unit (23) corrects the output voltage command value. The correction amount of the output voltage command value is
The control device for a power converter according to claim 2, wherein the control is performed so as to be larger than an increase amount of the output voltage calculated from a deviation between the output voltage command value and an actual value reduced by the correction of the output correction unit. . The circuit impedance used for calculating the reference voltage is
A resistance component R from the DC power supply to the capacitor;
An equivalent series resistance component Rc of the capacitor;
From the inductance component L, the resistance component R, the capacitor capacitance C, and the equivalent series resistance component Rc of the capacitor from the DC power supply to the capacitor,

Estimated impedance Z calculated by:
The control device for a power converter according to any one of claims 1 to 3, wherein the control value is a maximum value among the following.

Images