DE102011088915B4 - Calculation of a reverse induced residual current ripple at a DC input of a motor controller for a synchronous machine - Google Patents

Abstract

Method (100) for reducing a backward-induced residual current ripple at a DC input of a motor control device (216) for a synchronous machine (222), the method comprising - measuring (102) a phase current on a stator winding of the synchronous machine (222), - calculating ( 104) a DC current IDC(214), which has a residual current ripple, at the DC input of the motor control device (216) from the measured phase current, - determining (106) a time-dependent desired current value through the stator winding based on the DC current IDC(214 ), so that the backward-induced residual current ripple is reduced, characterized in that- the calculation (104) of the residual current ripple having DC current IDC(214) is also based on • first time durations tu, tv, tw, during which a current flows through the flows through the respective stator windings,• a second period of time tmin, during which current flows simultaneously through all the stator windings, and• a period of time Tpwm of pulse width modulation for actual values of the phase currents through the stator windings,- where the following applies:IDC≈Iu*(tu−tmin)/Tpwm+IV*( tv−tmin)/Tpwm +Iw*(tw−tmin)/Tpwm, where tmin= min(tu, tv, tw) and Iu, Iv, Iw are values of the measured phase currents and the determination (106) of the time-dependent current setpoint Id, Iq is determined at the stator windings in the d/q system according to Id=Id0+k1*sin(6ωt)+k2*cos(6ωt)Iq=Iq0−k2*sin(6ωt)+k1*cos(6ωt), where Id0 and Iq0 are direct current components of the are current commands, and ω is the angular frequency of the phase currents on the stator windings.

Description

The invention relates to a method for reducing a reverse-induced residual current ripple at a DC input of a motor control device for a synchronous machine. In addition, the invention relates to a motor control device for a synchronous machine.

In the ideal case, the stator windings of electrical AC machines are assumed to be spatially sinusoidally distributed over the stator. In order to reduce the manufacturing costs for stator windings of synchronous machines, so-called concentrated or so-called non-sinusoidal windings are often used. Through these backward-looking induction currents with spatially distributed harmonics are induced. Typically, the sixth harmonic of the fundamental of the alternating current of the synchronous machine is particularly pronounced. This generally leads to a current ripple or residual current ripple or ripple in general on a connecting line of a battery, for example in an electric or hybrid vehicle. In addition, there are irregularities in the torque of the synchronous motor.

Efforts have been made in the past to reduce the torque irregularities or current ripple. To this end, it has been proposed, among other things, to use a torque sensor or acceleration sensor in order to measure torque vibrations that occur and to take countermeasures. However, such a sensor adds cost and is therefore undesirable. Without a direct measurement of the irregularities of the torque, a model-based torque estimation is alternatively possible. However, this requires precise knowledge or an exact model of the synchronous machine and the resulting torque. This requires exact knowledge of the parameters of the synchronous machine. This knowledge is typically not available.

According to a similar method as in the torque irregularity reduction, DC ripple reductions on a connection from a battery to a motor controller for the synchronous machine can also be calculated when the DC current is measured, estimated or determined by an offline model. However, it is impossible to reduce the DC ripple or angular momentum irregularities at the same time in a permanent magnet synchronous machine (PSM). Reducing torque ripple is equivalent to “injecting harmonics” into a power controller to produce torque with no ripple, but creating backward-looking inductance with harmonics. However, if the current contains harmonics, the motor inductance and resistance will draw power at the frequency of the harmonics. In this case, the direct current will have a current ripple with harmonics.

In addition, it is possible to reduce the DC ripple effect by using a larger smoothing capacitor connected in parallel with the battery or in parallel with a high-current battery network. However, a larger capacitor also means higher costs and a higher space requirement. Both effects are undesirable in electric or hybrid vehicles.

The publication US 2011/0 238 245 A1 describes, for example, a method for reducing the current ripple of a DC current for supplying an electric motor.

Other publications such as WO 2011/128695 A2 , WO 2011/107773 A2 or a paper by GMS Azevedo, MC Cavalcanti, FAS Neves, LR Limongi, and KC Oliveira entitled "Grid Connected Photovoltaic Topologies with Current Harmonic Compensation" also addresses current ripple reduction.

The object of the invention is to minimize a DC ripple at the input of a control device for a synchronous machine without using a larger smoothing capacitor or without additional sensors.

This object is solved by the subject matter of the independent patent claims. Advantageous embodiments of the present invention are described in the dependent patent claims.

wobei t_min = min (t_u, t_v, t_w) und I_u, I_v, I_w Werte der gemessenen Phasenströme sind.According to a first aspect of the invention, a method for reducing a reverse-induced residual current ripple at a DC input of a motor control device for a synchronous machine is described. The method includes: measuring at least one phase current through a stator winding the synchronous machine; calculating a DC current having a residual current ripple at a DC input of the motor control device from the measured phase current; and determining a time-dependent desired current value through the stator winding based on the DC current having the residual current ripple so that the reverse-induced residual current ripple is reduced. The calculation of the DC current exhibiting the residual current ripple is also based on: first durations t _u , t _v , t _w during which a current flows through the respective stator windings, a second duration t _min , during which current flows simultaneously through all stator windings, and a time period T _PWM of pulse width modulation for actual values of the phase currents through the stator windings. The following applies: $$\begin{array}{l}{\text{I}}_{\text{DC}}\approx {\text{I}}_{\text{u}}*\left({\text{t}}_{\text{u}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{PWM}}+{\text{I}}_{\text{V}}*\left({\text{t}}_{\text{v}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{PWM}}\\ +{\text{I}}_{\text{W}}*\left({\text{t}}_{\text{W}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{PWM}},\end{array}$$

where t _min = min (t _u , t _v , t _w ) and I _u , I _v , I _w are values of the measured phase currents.

Advantageously, all three phase currents are measured at the respective stator windings, or two of the three phase currents are measured and the third phase current is determined from the other two phase currents measured. The DC current, which has a residual current ripple, at a DC input of the motor control device is then calculated from the three measured or determined phase currents.

The advantage of this method is that a residual current ripple that normally occurs on a power line from a battery in, for example, an electric or hybrid vehicle to an engine control device is reduced or completely eliminated. This reduction does not require a measurement--which would require an additional current sensor--of the input current of the motor control device, nor is it necessary to dimension a capacitor that is parallel to the input of the motor control device larger than was previously the case. In this way, both the costs for an additional current sensor and the costs or additional space for a larger smoothing capacitor can be avoided.

In this document, the term "residual current ripple" is used to describe a ripple in a direct current (DC current). This residual ripple can be caused by harmonics superimposed on the direct current. Such residual ripples are known for a direct current that has been rectified from an alternating current. In other cases, such ripples occur in currents from batteries on which converters for alternators are operated.

A backward-induced residual current ripple describes a ripple on a direct current, with the ripple arising from the fact that harmonics of a fundamental wave were superimposed on the direct current. This superimposition of reverse induced currents in a synchronous machine can occur with synchronous machines and associated converters that are not ideal or do not regulate ideally.

In this document, the term "synchronous machine" is intended to denote an electrical machine. Typically, synchronous machines have a rotating rotor with a winding powered by direct current and three static stators connected to the housing. Alternatively, the rotor can also have a permanent magnet. However, the person skilled in the art is also aware that a synchronous machine is fundamentally electromechanically symmetrical in relation to interchanging stators and rotors.

This term "motor control device" is to be understood here as meaning a control device for a synchronous machine. For example, within the motor controller, it is necessary to perform direct current to alternating current conversion, as well as current and voltage transformations between different frames. In addition, it should be assumed in this document that the engine control device also includes a high-voltage component and not just low-voltage-controlled control units are present. The engine control device is part of a so-called Power Electronic Block (PEB).

According to one exemplary embodiment, the calculation of a DC current that has the residual current ripple is also based on a degree of modulation of a modulation signal within the motor control device, which signal is used to define an operating point of the synchronous machine.

The method is thus advantageously based on signals, that is to say measured currents through the stator windings, and the degree of modulation of signals in the motor control device, which is known in the motor control device anyway.

According to a further exemplary embodiment, the determination of the time-dependent setpoint current value includes determining amplitudes of a harmonic in the calculated DC current having the residual current ripple. In this case, the harmonic corresponds to a fundamental of an electrical rotary frequency of the synchronous machine. In particular, the harmonic can be a sixth harmonic or harmonic of a rotary frequency of the synchronous machine.

The knowledge of the amplitudes of the harmonics on the ideally completely uniform DC current at the input of the motor control device is used to reduce or eliminate the reverse-induced residual current ripple - in particular the sixth harmonic.

According to a further exemplary embodiment, the determination of the time-dependent desired current value is also based on determining coefficients based on the determination of the amplitudes of the harmonics, so that the determination of the time-dependent desired current value is based on the coefficients.

This multi-stage method enables a particularly elegant feasibility for the calculation of the current that flows through the stator windings of the synchronous machine.

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{cos}\left(6\mathrm{\omega }\text{t}\right).$$

According to a further exemplary embodiment of the present invention, the determination of the time-dependent target current value through the stator windings in the d/q system is determined according to the following equations: $${\text{I}}_{\text{i.e}}={\text{I}}_{\text{i.e}0}+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_2*\text{cos}\left(6\mathrm{\omega }\text{t}\right)$$

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{cos}\left(6\mathrm{\omega }\text{t}\right).$$

I _d0 and I _q0 are the direct current components of the current setpoint and ω corresponds to the angular frequency of the phase currents through the stator windings.

As is known to the person skilled in the art, details for current values for synchronous machines can be given in the d/q system (direct-quadrature rotating reference frame). In this case, the knowledge of an angle between a rotating d/q reference system of the rotor and another reference system (eg the reference system of the stator) of the synchronous machine is used. In this context it should also be mentioned that a d/q transformation between the rotating frame of reference of the rotor and another frame of reference of the stator(s) (e.g. u, v, w) is a well-known mathematical transformation that is often used to generate a Perform analysis of three-phase electrical systems. In the case of three-phase, uniform, electrical circuits, an application of the d/q transformation to three alternating currents I _u , I _v , I _w means a reduction to two direct currents. As a result, calculations can be performed more easily on the basis of these imaginary direct currents I _d , I _q before they are converted back into the actual three-phase alternating currents by means of an inverse transformation.

According to a further aspect of the invention, a motor control device for a synchronous machine is specified. The motor control device has the following elements: current sensors for measuring at least one phase current through a stator winding of the synchronous machine and a control device. The control device in turn has the following elements: A DC current calculation unit, which is designed to calculate a DC current, which has a residual current ripple, at the DC input of the motor control device. The calculation is based on the at least one measured phase current through the stator winding of the synchronous machine. In addition, the control device has a setpoint current value calculation unit that is designed to determine a time-dependent setpoint current value through the stator winding based on the DC current that has the residual current ripple. This reduces a backward-induced residual current ripple at a DC input of the motor control device. The engine control device also has an extractor on the control device. This is designed to determine amplitudes of a harmonic in the DC current that has the residual ripple. In this case, the harmonic corresponds to a fundamental wave of an electrical rotary frequency of the synchronous machine. The harmonics are, in particular, sixth harmonics of the rotational frequency of the synchronous machine.

Advantages of this engine control device have already been discussed above in connection with the associated method.

dargestellt wird. ω ist dabei die Kreisfrequenz der Phasenströme durch die Statorwicklungen der Synchronmaschine. Dies ist insbesondere die sechste harmonische Oberschwingung der Rotationsfrequenz der Synchronmaschine. Der Extraktor führt dabei eine Fourier- bzw. Laplace-Transformation für eine einzelne Frequenz - hier die sechste Oberschwingung - aus.According to another embodiment, the extractor has a number of calculation elements. An output signal of the DC current calculation unit is present at a first positive input of a differential element. The extractor also has a first multiplier, at whose first signal input the output signal of the differentiator is present and at whose second signal input a signal cos(6ωT) is present. In addition, a first integration element—in particular using a convergence factor—is provided. The input of the integrator is connected to the output of the first multiplier. The output of the integrator is connected to a first input of a second multiplier. The signal cos(6ωt) is present at the second input of the second multiplier. Furthermore, a summation element is provided in the extractor. The first input of the summing element is connected to the output of the second multiplier. The output of the summation element is connected to the second negative input of the difference element. In addition, a third multiplier is provided, at whose first signal input the output of the differential element is present and at whose second signal input a signal sin (6ωt) is present. In addition, the extractor has a second integration element - in particular using the convergence factor mentioned above. The input of the second integrator is connected to the output of the third multiplier. The output of the integrator is connected to the first input of a fourth multiplier, with the signal sin(6ωt) being present at a second input of the fourth multiplier. The value of a signal aa can be brought out at the output of the first integration element and the value of a signal bb can be brought out at the output of the second integration element. After a transient, the value of the signal aa is identical to a partial amplitude a of the 6th harmonic of the DC current, and the value of the signal bb is identical to another partial amplitude b of the 6th harmonic of the DC current, where an output of the DC current calculation unit for the 6th harmonic as $${\text{I}}_{\text{DC}\left(6\text{th}\right)}=\text{a}*\text{cos}\left(6\mathrm{\omega }\text{t}\right)+\text{b}*\text{sin}\left(6\mathrm{\omega }\text{t}\right)$$

is pictured. ω is the angular frequency of the phase currents through the stator windings of the synchronous machine. In particular, this is the sixth harmonic of the rotational frequency of the synchronous machine. The extractor carries out a Fourier or Laplace transformation for a single frequency - here the sixth harmonic.

According to a further exemplary embodiment, the control device also has a coefficient determiner, which is designed to determine coefficients from the amplitudes a and b of the harmonic, which in turn are input variables for the current setpoint value calculation unit.

In this way, residual ripples in the input current of a motor control device for a synchronous machine can be reduced or completely eliminated in a very elegant way, without the need for additional metrological sensors or larger smoothing capacitors. Other electrical loads that are connected to the DC current network in parallel with the motor control device are therefore no longer affected by the motor control device.

It is pointed out that embodiments of the invention have been described with reference to different objects of the invention. In particular, some embodiments of the invention are described with apparatus claims and other embodiments of the invention are described with method claims. However, it will be immediately clear to those skilled in the art upon reading this application that, unless explicitly stated otherwise, any combination of features belonging to different types of subject matter is also possible in addition to a combination of features belonging to one type of subject matter objects of the invention.

Advantageous refinements of the method presented above are also to be regarded as advantageous refinements of the motor control device, insofar as they can otherwise be transferred to the motor control device presented above.

• 1 zeigt ein Blockdiagramm des erfindungsgemäßen Verfahrens.
• 2 zeigt ein Blockschaltbild eines Stromkreises mit einer Batterie einer Motorsteuervorrichtung und einer Synchronmaschine.
• 3 zeigt ein Blockdiagramm des erfindungsgemäßen Verfahrens in größerem Detail.
• 4 zeigt ein Blockdiagramm des Extraktors.
• 5 zeigt eine Restwelligkeit im Drehmoment einer Synchronmaschine im Vergleich zu einer Restwelligkeit des DC-Eingangsstromes der Motorsteuervorrichtung.
• 6 zeigt eine alternative Reduktion einer Restwelligkeit des Drehmoments einer Synchronmaschine im Vergleich zur Restwelligkeit des DC-Eingangsstromes der Motorsteuervorrichtung.

Further advantages and features of the present invention result from the following exemplary description of currently preferred embodiments. The individual figures of the drawing of this application are to be regarded as merely schematic and not true to scale.

• 1 shows a block diagram of the method according to the invention.
• 2 shows a block diagram of a circuit with a battery of a motor control device and a synchronous machine.
• 3 shows a block diagram of the method according to the invention in more detail.
• 4 shows a block diagram of the extractor.
• 5 12 shows a torque ripple of a synchronous machine compared to a DC input current ripple of the motor controller.
• 6 shows an alternative reduction of a residual ripple of the torque of a synchronous machine compared to the residual ripple of the DC input current of the motor control device.

It is pointed out that features or components of different embodiments that are the same or at least functionally the same as the corresponding features or components of the embodiment are provided with the same reference numbers or with a different reference number, which differs only in its first digit from distinguishes the reference sign of a (functionally) corresponding feature or a (functionally) corresponding component. To avoid unnecessary repetition, features or components that have already been explained using a previously described embodiment will not be explained in detail later. Furthermore, it is pointed out that the embodiments described below only represent a limited selection of possible embodiment variants of the invention. In particular, it is possible to combine the features of individual embodiments with one another in a suitable manner, so that a large number of different embodiments can be regarded as obviously disclosed for the person skilled in the art with the embodiment variants explicitly presented here.

As a mathematical background to the invention, reference is made to the following mathematical model for a synchronous machine with a permanent magnet (PSM=permanent magnet synchronous machine) as the rotor. In principle, the invention can also be used in connection with an externally excited synchronous machine.

In the case of non-sinusoidal backward-facing inductions and with a transformation of the voltage vector into rotor coordinates, the harmonics on the supplied direct current consist in particular of sixth harmonics of a fundamental oscillation of the synchronous machine.

$${\mathrm{\Psi }}_q=L_q\cdot I_q+{\mathrm{\Psi }}_{PM}\cdot \left[\frac{7\cdot a5-5\cdot a7}{35}\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\right]$$

$$u_d=R_s{I}_d+\frac{d{\mathrm{\Psi }}_d}{dt}-\omega \cdot {\mathrm{\Psi }}_q$$

$$u_q=R_s{I}_q+\frac{d{\mathrm{\Psi }}_q}{dt}+\omega \cdot {\mathrm{\Psi }}_d$$

$$M=\frac{3}{2}\cdot P_p\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_d-L_q\right)\cdot I_d\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\end{array}\right\}$$

$$\approx \frac{I_{dc}\cdot V_{dc}-P_{impendance}}{{\omega }_{mech}}$$

$$I_{dc}=\frac{3}{2}\cdot P_p\cdot k_{dc}\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_d-L_q\right)\cdot I_d\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\end{array}\right\}\angle {\theta }_{dc\_M}$$

From this, the machine can be modeled with a permanent magnet with sixth harmonics as follows: $${\mathrm{\Psi }}_{\mathrm{i.e}}=L_{\mathrm{i.e}}\cdot I_{\mathrm{i.e}}\cdot +{\mathrm{\Psi }}_{PM}\left[1-\frac{7\cdot a5+5\cdot a7}{35}\cdot \text{cos}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\right]$$

$${\mathrm{\Psi }}_q=L_q\cdot I_q+{\mathrm{\Psi }}_{PM}\cdot \left[\frac{7\cdot a5-5\cdot a7}{35}\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\right]$$

$${\mathrm{and}}_{\mathrm{i.e}}=R_s{I}_{\mathrm{i.e}}+\frac{\mathrm{i.e}{\mathrm{\Psi }}_{\mathrm{i.e}}}{\mathrm{i.e}t}-\omega \cdot {\mathrm{\Psi }}_q$$

$${\mathrm{and}}_q=R_s{I}_q+\frac{\mathrm{i.e}{\mathrm{\Psi }}_q}{\mathrm{i.e}t}+\omega \cdot {\mathrm{\Psi }}_{\mathrm{i.e}}$$

$$M=\frac{3}{2}\cdot P_p\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}}\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\end{array}\right\}$$

$$\approx \frac{I_{\mathrm{i.e}c}\cdot V_{\mathrm{i.e}c}-P_{impen\mathrm{i.e}ance}}{{\omega }_{mecH}}$$

$$I_{\mathrm{i.e}c}=\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}}\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\left({\theta }_{el}+\frac{\pi }{2}\right)\right)\end{array}\right\}\angle {\theta }_{\mathrm{i.e}c\_M}$$

ΨPM

Flussverkettung des Permanentmagneten

Ld, Lq

Motorinduktivität

θel

elektrischer Winkel

Pp

Anzahl der Polpaare

IDC

DC-Strom am Gleichstromanschluss

ω

elektrische Winkelgeschwindigkeit

ωmech

mechanische Winkelgeschwindigkeit

Pimpedance

von der Motorimpedanz verbrauchte Leistung

kdc

Transformationsfaktor zwischen Drehmoment und DC-Strom

θdc_M

Phasenwinkel zwischen Drehmoment und DC-Strom

as, a7

Komponentenamplituden der fünften und siebten harmonischen Oberwelle bezogen auf Stator-Koordinaten, wenn die Grundfrequenz gleich 1 ist.

The following applies:

ΨPM

Flux linkage of the permanent magnet

Ld, Lq

motor inductance

θel

electrical angle

pp

number of pole pairs

I.D.C

DC current on the DC connector

ω

electrical angular velocity

ωmech

mechanical angular velocity

pimp dance

power dissipated by the motor impedance

kdc

Transformation factor between torque and DC current

θdc_M

Phase angle between torque and DC current

ace, a7

Component amplitudes of the fifth and seventh harmonics referred to stator coordinates when the fundamental frequency is 1.

dann ergibt sich zur Elimination z.B. der sechsten harmonischen Oberwelle des Drehmoment-Ripples oder eines DC-Strom-Ripples $${\text{I}}_{\text{d}}={\text{I}}_{\text{d}0}+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right),$$

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right).$$

In the event that I _d and I _q target values are given together with sixth harmonics instead of constant values and if it is defined that $${\theta }_{el}+{\theta }_{\mathrm{i.e}c\_M}+\mathrm{\pi }\text{/}2=\mathrm{\omega }\text{t},$$

then, for example, the elimination of the sixth harmonic of the torque ripple or a DC current ripple results $${\text{I}}_{\text{i.e}}={\text{I}}_{\text{i.e}0}+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right),$$

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right).$$

Here, k ₁ and k ₂ are coefficients that must be determined. They are functions of the engine parameters. Using the calculated I _d and I _q as setpoints minimizes the sixth harmonics of DC current ripple or torque ripple. In the event that the DC current ripple is to be minimized, the value of the associated signal is required and I _d and I _q are adjusted until the DC current ripple is minimized. If a signal that represents the torque ripple is used instead of the DC current signal, the torque ripple is minimized.

Correct machine parameters are required for determining correct coefficients k ₁ and k ₂ . However, this would be difficult or even impossible for each operating point of the synchronous machine.

Instead, an approximation method using the intensities of the harmonics in the DC current can be used to determine the coefficients.

$$I_q=I_{q0}-k_1\cdot \text{sin}\left(6\omega t\right)+k_1\cdot \text{cos}\left(6\omega t\right)=I_{q0}+\mathrm{\Delta }{I}_q$$

Suppose you can set the current like this: $$I_{\mathrm{i.e}}=I_{\mathrm{i.e}0}+k_1\cdot \text{sin}\left(6\omega t\right)+k_2\cdot \text{cos}\left(6\omega t\right)=I_{\mathrm{i.e}0}+\mathrm{\Delta }{I}_{\mathrm{i.e}}$$

$$I_q=I_{q0}-k_1\cdot \text{sin}\left(6\omega t\right)+k_1\cdot \text{cos}\left(6\omega t\right)=I_{q0}+\mathrm{\Delta }{I}_q$$

$$$$

$$$$

$$+\frac{3}{2}\cdot P_p\cdot k_{dc}\cdot \underset{12.harmonischeOberwelle}{\underbrace{\left[\begin{array}{c}\left(L_d-L_q\right)\cdot \mathrm{\Delta }{I}_d\cdot \mathrm{\Delta }{I}_q\\ +A_2\text{ cos}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot \mathrm{\Delta }{I}_q\\ -A_1\text{ sin}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot \mathrm{\Delta }{I}_d\end{array}\right]}}$$

Then from the DC current equation it is known: $$\begin{array}{l}{I}_{\mathrm{i.e}c}=\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}}\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\omega t\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\omega t\right)\end{array}\right\}\\ =\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\cdot \left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot I_{q0}\\ +\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\cdot \underset{6.Ha\mathrm{right}mOniscHeObe\mathrm{right}welle}{\underbrace{\left\{\begin{array}{l}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot \mathrm{\Delta }{I}_q+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\cdot \mathrm{\Delta }{I}_{\mathrm{i.e}}\\ +A_2\text{cos}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot I_{q0}-A_1\text{sin}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot I_{\mathrm{i.e}0}\end{array}\right\}}}\end{array}$$

$$$$

$$$$

$$+\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\cdot \underset{12.Ha\mathrm{right}mOniscHeObe\mathrm{right}welle}{\underbrace{\left[\begin{array}{c}\left(L_{\mathrm{i.e}}-L_q\right)\cdot \mathrm{\Delta }{I}_{\mathrm{i.e}}\cdot \mathrm{\Delta }{I}_q\\ +A_2\text{cos}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot \mathrm{\Delta }{I}_q\\ -A_1\text{sin}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot \mathrm{\Delta }{I}_{\mathrm{i.e}}\end{array}\right]}}$$

The following applies: $${\text{A}}_1=a_5+a_7,{\text{A}}_2=a_5-a_7$$

If one neglects the 12th harmonic and if one can force the sixth harmonic to be zero, then the 6th harmonic of the DC current is given by: $$\begin{array}{l}{I}_{\mathrm{i.e}c}=\frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}}\right]\cdot I_q\\ +{\mathrm{\Psi }}_{PM}\cdot \left(a5-a7\right)\cdot \text{cos}\left(6\omega t\right)\\ -{\mathrm{\Psi }}_{PM}\cdot \left(a5+a7\right)\cdot \text{sin}\left(6\omega t\right)\end{array}\right\}\\ \approx \frac{3}{2}\cdot P_p\cdot k_{\mathrm{i.e}c}\cdot \left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot I_{q0}\\ +\frac{3}{2}\cdot P_p{k}_{\mathrm{i.e}c}\cdot \underset{6.Ha\mathrm{right}mOniscHeObe\mathrm{right}welle\to 0}{\underbrace{\left\{\begin{array}{c}\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot \left[k_1\text{cos}\left(6\omega t\right)-k_2\text{sin}\left(6\omega t\right)\right]\\ +\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\cdot \left[k_2\text{cos}\left(6\omega t\right)-k_1\text{sin}\left(6\omega t\right)\right]\\ +A_2\text{cos}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot I_{q0}-A_1\text{sin}\left(6\omega t\right)\cdot {\mathrm{\Psi }}_{PM}\cdot I_{\mathrm{i.e}0}\end{array}\right\}}}\end{array}$$

For cos(6ωt) applies $$\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot k_1+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot k_2=-A_2\cdot {\mathrm{\Psi }}_{PM}\cdot I_{q0}$$

For sin(6ωt) applies $$\left[{\mathrm{\Psi }}_{PM}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot \left(-k_2\right)+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot k_1=A_1\cdot {\mathrm{\Psi }}_{PM}\cdot I_{\mathrm{i.e}0}.$$

From these two equations k ₁ and k ₂ can be determined. Instead of solving these algebraic equations “offline”, it is possible to determine the coefficients “online” by an approximation algorithm:

$${\epsilon }_2=\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_d-L_q\right)\cdot I_{d0}\right]\cdot \left(-k_2\right)+\left(L_d-L_q\right)\cdot I_{q0}\cdot k_1+A_1\cdot {\mathrm{\Psi }}_{P{M}_0}\cdot I_{d0}$$

The errors ε ₁ and ε ₂ are defined as follows: $$e_1=\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot k_1+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot k_2+A_2\cdot {\mathrm{\Psi }}_{P{M}_0}\cdot I_{q0}$$

$$e_2=\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot \left(-k_2\right)+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot k_1+A_1\cdot {\mathrm{\Psi }}_{P{M}_0}\cdot I_{\mathrm{i.e}0}$$

• Die sechsten harmonischen Oberwellen des DC-Strom-Ripple lassen sich als

$$I_{dc\left(6th\right)}=a\cdot \text{cos}\left(6\omega t\right)+b\cdot \text{sin}\left(6\omega t\right)=\gamma \left[{\epsilon }_1\cdot \text{cos}\left(6\omega t\right)+{\epsilon }_2\cdot \text{sin}\left(6\omega t\right)\right]$$

darstellen. Wobei $$\text{a}=\mathrm{\gamma }*{\mathrm{\epsilon }}_1\text{ und a}=\mathrm{\gamma }*{\mathrm{\epsilon }}_2\text{ gilt}.$$

These are proportional to the amplitudes of the sixth harmonic of the DC current. a and b still have a factor γ. So:

• The sixth harmonics of the DC current ripple can be expressed as

$$I_{\mathrm{i.e}c\left(6tH\right)}=a\cdot \text{cos}\left(6\omega t\right)+b\cdot \text{sin}\left(6\omega t\right)=g\left[e_1\cdot \text{cos}\left(6\omega t\right)+e_2\cdot \text{sin}\left(6\omega t\right)\right]$$

represent. Whereby $$\text{a}=\mathrm{g}*{\mathrm{e}}_1\text{and a}=\mathrm{g}*{\mathrm{e}}_2\text{is applicable}.$$

If the correct k ₁ and k ₂ are found, the error is reduced to zero. This also reduces the DC current ripple to zero. Knowledge of γ is therefore not required.

$$\frac{d{k}_2}{dt}=-\alpha \cdot \left\{-\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_d-L_q\right)\cdot I_{d0}\right]\cdot b+\left(L_d-L_q\right)\cdot I_{q0}\cdot a\right\}$$

Thus will $$\frac{\mathrm{i.e}{k}_1}{\mathrm{i.e}t}=-a\cdot \left\{\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot a+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot b\right\}$$

$$\frac{\mathrm{i.e}{k}_2}{\mathrm{i.e}t}=-a\cdot \left\{-\left[{\mathrm{\Psi }}_{P{M}_0}+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{\mathrm{i.e}0}\right]\cdot b+\left(L_{\mathrm{i.e}}-L_q\right)\cdot I_{q0}\cdot a\right\}$$

Where α is the convergence factor that determines the convergence speed with respect to k ₁ and k ₂ . Thus the entire convergence dynamics is given by k (see in connection with 4 ) and α determined.

1 10 shows a block diagram of the method 100 for reducing a reverse-induced residual current ripple at a DC input of a motor control device for a synchronous machine according to an embodiment. The method has: measuring 102 phase currents through the stator windings of the synchronous machine, calculating 104 a DC current at the DC input of the motor control device that has a residual current ripple from the measured phase currents, determining 106 a time-dependent desired current value through the stator windings based on the DC current that has the residual current ripple current so that reverse induced residual current ripple is reduced.

2 shows a block diagram of an electric circuit with a battery 202, a motor control device 216 and a synchronous machine 222. The battery or high-voltage battery 202 is shown as an equivalent circuit diagram with an internal resistance 204 and an inductance 206. The battery is connected to a high-voltage power network 208 . Motor control device 216 is in turn connected to this high-voltage power supply system. At point 210 the supply voltage of the high-voltage battery is therefore present at the high-voltage control block 224 (PEB=Power Electric Block). Reference number 214 represents the current that would be measurable at the input of motor controller 216 . A smoothing capacitor 212 limits a current ripple of the current 214 . A synchronous machine 222 is connected to the motor control device 216 . In this exemplary embodiment, three connecting lines 220, which are intended to represent connections to the stator windings, are drawn in. Typically, the currents through lines 220 are measured. Currents induced backwards occur at the high-voltage input of motor control device 216 as a result of non-ideal control in motor control device 216 or non-ideal windings of synchronous machine 222 . These need to be reduced. One possibility for a further reduction would be to increase the capacitance of the capacitor 212, which, however, would be associated with higher costs and a higher space requirement on the PEB 224. An alternative would be to measure the current 214, although this would require an additional current sensor. This is because such a current sensor requires additional assembly work and additional costs for the control of the synchronous machine. Such costs should be avoided as far as possible.

3 illustrates the procedure in a greater degree of detail 3 simultaneously represents an implementation of an embodiment of the invention from corresponding functional components or calculation units. First, by means of unit 302, the direct current I _DC at the input of Engine control device calculated. This calculation is based on measured current values I _u,v,w through lines 220 (cf. 2 ). Other input variables for this calculation are t _min , which corresponds to a period of time during which current flows through all stator windings at the same time. The time period t _PVM is the time period (total time period) of pulse width modulation for actual values of the phase currents through the stator windings. The input variables t _u,v,w are durations during which current flows through the associated (u,v,w) stator windings. The result of this calculation of I _DC is used as an input signal for an extractor 304 . As a result, the extractor 304 provides two signals: a and b, which represent amplitudes of the current I _DC with reference number 214 including an undesired residual current ripple. Based on these amplitudes a and b, in block 306 coefficients k ₁ and k ₂ are calculated. These values k ₁ and k ₂ in turn serve as a basis for calculating current setpoints, which are shown as I _d setpoint or I _q setpoint in the d/q system and are calculated by a current setpoint calculation unit 308 .

4 shows the extractor in greater detail. The calculated I _DC ( 2 , 214). This has a harmonic that can also be represented in the form of a*cos(6ωt) + b*sin(6ωt). Here ω is the electrical angular frequency of the synchronous machine. The Extractor ( 3 , 304) executes a Fourier transformation or a Laplace transformation, as it were. A specific pronounced frequency corresponds to the sixth harmonic of the fundamental frequency of the synchronous machine. The extractor has a difference element 402, four multiplication units 404, 408, 410 and 414. In addition, two integration units 406 and 412 are provided. The outputs of the two multipliers 408 and 414 are fed to a summing element 416 . The output of the summation element 416 is fed back to the negative input of the difference element 302. The individual components are in accordance with the circuit diagram 4 merged. It can be seen that signal outputs are provided at the outputs of the integration elements 406 and 412 . The signal outputs are labeled aa and bb. A convergence factor k of the integrators 406 and 412 is typically determined experimentally in each case. Alternatively, a correct mathematical calculation would be possible. However, this would require a complete model of the behavior of the synchronous machine as well as the complete control mechanism for the synchronous machine. To avoid this effort, the convergence factor k of the extractor is usually determined experimentally. It is designed in such a way that after only a few passages of the signals through the extractor, the values aa correspond to the amplitude a and bb to the amplitude b of the DC current at the input of the motor control device, which has the residual current ripple.

The present method and the motor control device described are therefore suitable for reducing or eliminating the residual ripple of a DC input current at the motor control device. The basis for this is a phase current measurement and a PWM. If a torque measurement signal were available, the method described could alternatively also be used to eliminate a torque ripple.

5a shows the time course of a torque 502 of a synchronous machine, in which the ripple of the torque of the synchronous machine can be reduced only slightly. On the other hand, the residual current ripple 504 is reduced to a minimum after a short settling time, as described above ( 5b) .

Alternatively, in 6a the reduction of the ripple of the torque 602 of a synchronous machine is shown on the basis of an exemplary embodiment of the present invention. One can clearly see that after a short settling time, the ripple of the torque of the synchronous machine is significantly reduced. However, how will look like 6b As can be seen, at the same time the residual ripple in the DC input current 604 of the motor control device is slightly increased. The 5 and 6 represent the results of simulation calculations that take the inventive method as a basis.

In summary, it remains to be stated that a residual current ripple induced backwards at a DC input of a motor control device for a synchronous machine can be elegantly reduced or eliminated. For this purpose it is not necessary to increase the capacity of a smoothing capacitor at the input of the motor control device or it is not necessary to provide an additional current sensor at the input of the motor control device for measuring the input DC current. The residual current ripple of the current I _DC 214 can be reduced very cost-effectively using the method described. This avoids an additional load on the high-voltage battery 202 or a load on or impairment of other electrical or electronic devices that are connected to the high-voltage network 208 .

Claims (7)

Method (100) for reducing a backward-induced residual current ripple at a DC input of a motor control device (216) for a synchronous machine (222), the method comprising - measuring (102) a phase current at a stator winding of the synchronous machine (222), - calculating ( 104) a DC current I _DC (214) having a residual current ripple at the DC input of the motor control device (216) from the measured phase current, - determining (106) a time-dependent desired current value through the stator winding based on the DC current I _DC having the residual current ripple (214) so that the backward-induced residual current ripple is reduced, characterized in that - the calculation (104) of the residual current ripple having DC current I _DC (214) is also based on • first time durations t _u , t _v , t _w , during in which a current flows through the respective stator windings, • a second time period t _min during which current flows simultaneously through all stator windings, and • a time period T _pwm of a pulse width modulation for actual values of the phase currents through the stator windings, - where the following applies: $$\begin{array}{l}{\text{I}}_{\text{DC}}\approx {\text{I}}_{\text{and}}*\left({\text{t}}_{\text{and}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}}+{\text{I}}_{\text{V}}*\left({\text{t}}_{\text{v}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}}\\ +{\text{I}}_{\text{w}}*\left({\text{t}}_{\text{w}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}},\end{array}$$

where t _min = min(t _u , t _v , t _w ) and I _u , I _v , I _w are values of the measured phase currents and the determination (106) of the time-dependent desired current value I _d , I _q at the stator windings in d/ q system according to $${\text{I}}_{\text{i.e}}={\text{I}}_{\text{i.e}0}+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_2*\text{cos}\left(6\mathrm{\omega }\text{t}\right)$$

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{cos}\left(6\mathrm{\omega }\text{t}\right)$$

is determined, where I _d0 and I _q0 are direct current components of the desired current values, and ω is the angular frequency of the phase currents at the stator windings. Method (100) according to claim 1 , wherein the calculation (104) of the residual current ripple having DC current I _DC (214) is also based on a degree of modulation of a modulation signal within the motor control device (216) to define an operating point of the synchronous machine (222). Method (100) according to claim 1 or 2 , wherein the determination (106) of the time-dependent desired current value comprises a determination (304) of amplitudes of a harmonic in the calculated DC current I _DC (214) having the residual current ripple, the harmonic corresponding to a fundamental of an electrical rotary frequency of the synchronous machine (222). . Method (100) according to claim 3 , wherein the determination (106) of the time-dependent desired current value additionally comprises a determination (306) of coefficients based on the determination (304) of the amplitudes of the harmonics, so that the determination of the time-dependent desired current value is based on the determined coefficients. Motor control device (216) for a synchronous machine (222), the motor control device (216) having - current sensors for measuring a phase current at a stator winding of the synchronous machine, and a control device (218), the control device having: • a DC current calculation unit ( 302) which is designed to calculate a DC current (214) at the DC input of the motor control device (216) which has a residual current ripple, the calculation being based on the measured phase current (220) through the stator winding of the synchronous machine (222), and • a current setpoint calculation unit (308), which is designed to determine a time-dependent current setpoint through the stator winding based on the DC current (214) having the residual current ripple, so that a backward-induced residual current ripple at a DC input of the motor control device (216) is reduced, characterized in that the control device (218) further comprises an extractor (304) which is designed to determine amplitudes of a harmonic in the DC current (214) having the residual ripple, the harmonic being a fundamental wave of an electrical rotation frequency of the synchronous machine (222), wherein - the DC current calculation unit (302) is set up in such a way that the calculation (104) of the residual current ripple having DC current I _DC (214) is also based on • first time durations t _u , t _v , t _w , during which a current flows through the respective stator windings, • a second time duration t _min during which current flows simultaneously through all stator windings, and • a time period T _pwm of pulse width modulation for actual values of the phase currents through the stator windings, - where is applicable: $$\begin{array}{l}{\text{I}}_{\text{DC}}\approx {\text{I}}_{\text{and}}*\left({\text{t}}_{\text{and}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}}+{\text{I}}_{\text{v}}*\left({\text{t}}_{\text{v}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}}\\ +{\text{I}}_{\text{w}}*\left({\text{t}}_{\text{w}}-{\text{t}}_{\text{at least}}\right){\text{/T}}_{\text{pwm}},\end{array}$$

where t _min = min(t _u , t _v , t _w ) and I _u , I _v , I _w are values of the measured phase currents and the current setpoint calculation unit (308) is set up to determine (106) the time-dependent current setpoint I _d , I _q on the stator windings in the d/q system according to $${\text{I}}_{\text{i.e}}={\text{I}}_{\text{i.e}0}+{\text{k}}_1*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_2*\text{cos}\left(6\mathrm{\omega }\text{t}\right)$$

$${\text{I}}_{\text{q}}={\text{I}}_{\text{q}0}-{\text{k}}_2*\text{sin}\left(6\mathrm{\omega }\text{t}\right)+{\text{k}}_1*\text{cos}\left(6\mathrm{\omega }\text{t}\right),$$

where I _d0 and I _q0 are DC components of the current setpoints, and ω is the angular frequency of the phase currents on the stator windings. Motor control device (216) according to claim 5 , the extractor (304) having the following elements: - a differential element (402), at whose first positive input an output signal (I _DC ) of the DC current calculation unit (302) is present, - a first multiplier (404), at whose the output signal of the differentiator (402) is present at the first signal input and a signal cos(6ωt) is present at its second signal input, - a first integrator (406) whose input is connected to the output of the first multiplier (404) and whose output is connected to the first input of a second multiplier (408), the signal cos(6ωt) being present at a second input of the second multiplier (408), - a summing element (416) whose first input is connected to the output of the second multiplier (408). and whose output is connected to the second negative input of the differential element (402), - a third multiplier (410), at whose first signal input the output signal of the differential element (402) is present and at whose second signal input a signal sin(6ωt) is present, - a second integrator (412) whose input is connected to the output of the third multiplier (410) and whose output is connected to the first input of a fourth multiplier (414), the signal sin(6ωt) being present at a second input of the fourth multiplier is present, and a value of a signal aa is output at the output of the first integration element (406) and a value of a signal bb is output at the output of the second integration element (412), so that after a transient process the value of the signal aa is identical to the amplitude a of the harmonic and the value of the signal bb is identical to the amplitude b of the sixth harmonic, an output signal I _DC of the current calculation unit being taken as $${\text{I}}_{\text{DC}}=\text{a}*\text{cos}\left(6\mathrm{\omega }\text{t}\right)+\text{b}(\text{sin}\left(6\mathrm{\omega }\text{t}\right)$$

and where ω is the angular frequency of the phase currents on the stator windings. The engine control device according to claim 5 or 6 , wherein the control device (218) further comprises a coefficient determiner (306) which is designed to determine coefficients from the amplitudes of the harmonics, which are input variables for the current setpoint calculation unit (308).

Images