Method for operating an electric motor

Abstract

The invention relates to a method for operating an electric motor (4) by means of pulse width modulation, in which, during a period duration (Tperiode), at least one pulse width modulated signal pulse (70) is generated at a pulse position (T0) with a pulse duration (Ton), wherein a resulting amplitude spectrum (64) has an amplitude value (A) at a critical frequency (ꞷkritisch), wherein at least one time parameter (Tperiode, T0, Ton) of the pulse width modulation is replaced by a sum of the time parameter (Tperiode, T0, Ton) and a correction factor (Tkorr), wherein the correction factor (Tkorr) is changed in such a way that a calculated amplitude spectrum (64') at the critical frequency (ꞷkritisch) has a reduced amplitude value (Iamp), and wherein the pulse width modulation is performed with the time parameter (Tperiode, T0, Ton) and the modified correction factor (Tkorr).

Description

[0001] Description

[0002] Method for operating an electric motor

[0003] The invention relates to a method for operating an electric motor using pulse-width modulation, in which at least one pulse-width modulated signal pulse is generated during a period. The invention further relates to an electric machine with an electric motor and software on a data carrier.

[0004] Electric motor-driven or -operated adjustment systems as automotive components, such as window regulators, seat adjusters, door and sunroof drives, or radiator fan drives, as well as pumps and interior blowers, typically feature an electric drive with a controlled electric motor. Brushless electric motors (brushless direct current motors, BLDC motors) are increasingly being used for such electric drives. These motors replace the wear-prone brush elements of a rigid (mechanical) commutator with electronic commutation of the motor current.

[0005] Electric motor drives in motor vehicles are typically powered by a (high-voltage) battery, which serves as the vehicle's internal energy storage device. This energy storage device supplies the electric motor with electrical energy in the form of direct current (or direct voltage) via the vehicle's electrical system. To convert this direct current into alternating current suitable for the motor, a power converter (inverter) is connected between the energy storage device and the electric motor. The power converter has a bridge circuit that is connected to the direct current or direct voltage supply of the energy storage device via an electrical intermediate circuit. The motor current is generated by pulse-width modulation (PWM) of the semiconductor switches in the bridge circuit, which delivers a multi-phase output current (three-phase current).

[0006] With PWM, the width (duration) of the voltage pulses is varied to control the average voltage and thus the power delivered to the motor. This is achieved through a predefined timing scheme in which the semiconductor switches alternate between a conducting and a blocking state in a rhythmic pattern, enabling efficient and precise control of motor power.

[0007] The switching operations of the semiconductor switches during PWM generate pulsed currents in the intermediate circuit or the vehicle electrical system. These currents or current ripples can lead to electromagnetic compatibility (EMC) disturbances. These EMC disturbances manifest themselves as unwanted electromagnetic interference, which can impair the normal function of other electronic devices or systems in the vehicle electrical system.

[0008] To meet EMC requirements for the electric motor or electric motor drive, current ripple is reduced, for example, through filtering, particularly using capacitors and / or inductors. Such filtering therefore requires additional components, resulting in increased costs and increased installation space requirements. Furthermore, it may happen that individual frequencies of the current ripple are not sufficiently dampened by the filtering. Especially if the filter resonates with the vehicle electrical system, EMC requirements may be exceeded at certain points.

[0009] The invention is based on the object of providing a particularly suitable method for operating an electric motor using pulse width modulation. In particular, the effort and costs required to comply with EMC requirements are to be reduced through targeted control of the pulse width modulation. The invention is further based on the object of providing a particularly suitable electrical machine for implementing the method and particularly suitable software on a data carrier.

[0010] The method according to the invention is intended for operating an electric motor using PWM, and is suitable and designed for this purpose. The electric motor is preferably part of a motor vehicle and is connected to an intermediate circuit of a (vehicle) electrical system.

[0011] During PWM, at least one pulse-width modulated signal pulse (PWM pulse) is generated at a pulse position with a pulse duration during a period.

[0012] A “period” is understood here and in the following to mean in particular the period of a complete PWM cycle repetition.

[0013] A “pulse duration” (pulse width, pulse length, pulse width) is understood here and in the following to mean in particular a period of time in which the PWM pulse is kept at a high (or low) level within the period duration.

[0014] A "pulse position" here and below refers in particular to the specific placement of the PWM pulse within the period, i.e., the temporal position of the PWM pulse relative to the beginning of the period. The pulse position refers, for example, to the middle of the PWM pulse. Alternatively, the pulse position can also refer to the beginning or end of the PWM pulse.

[0015] An amplitude spectrum resulting from PWM, in particular the amplitude spectrum of the resulting EMC interference or current ripple in an intermediate circuit or vehicle electrical system connected to the electric motor, exhibits an amplitude value at at least one critical frequency that does not meet specified EMC requirements. This means, for example, that the amplitude spectrum of the EMC interference caused by PWM in the intermediate circuit or vehicle electrical system is monitored and recorded, and that the process is initiated, for example, if the amplitude value does not meet the specified EMC requirements.

[0016] An “amplitude spectrum” is understood here and in the following to mean in particular the magnitude of a frequency spectrum.

[0017] A "critical frequency" is understood here and below to mean a specific frequency or frequency range in the amplitude spectrum of the current ripple at or within which amplitude values ​​of EMC interference occur that exceed specified EMC requirements or are at least considered critical in this regard. The critical frequency depends on various factors, such as the structure of the PWM signals, the type and quality of the electronic components involved, and the specific application conditions.

[0018] According to the method, at least one time parameter of the pulse width modulation is replaced by a sum of the time parameter and a correction factor. The time parameter is thus supplemented or modified with a correction factor.

[0019] A "time parameter" is understood here and below to mean a quantifiable measure of time that characterizes the essential aspects of the PWM process. These time parameters include, in particular, the period duration, the pulse duration, and the pulse position within the period duration.

[0020] A “correction factor” is understood to be a time variable that can be set or changed during the process, which is added to the time parameter and with which the pulse width modulation or the resulting amplitude spectrum is influenced.

[0021] With regard to the method according to the invention, the time parameter is a fixed temporal variable, which is supplemented by the correction factor as a variable temporal variable. The correction factor is changed such that a calculated amplitude spectrum has a reduced amplitude value at the critical frequency. PWM is then performed using the time parameter and the correction factor. This realizes a particularly suitable method for operating an electric motor using PWM. In particular, the EMC behavior of the electric motor is improved without the need for additional cost- and space-intensive electrical components.

[0022] The invention is based on the finding that the current ripple of the EMC interference has an associated amplitude spectrum which is attributable to the PWM or which can be changed or influenced by the PWM. According to the invention, no energy is removed from the amplitude spectrum by the correction factor; the spectrum, or rather the peak characterised by the critical frequency and the amplitude value, is merely (at least partially) frequency-shifted by the correction factor. In particular, the peak is shifted towards frequencies which are not critical with regard to EMC requirements. According to the invention, the energy of the current ripple is therefore shifted, whereby, together with filtering, the total EMC energy can also change.

[0023] Preferably, a predicted amplitude spectrum is calculated for a number of correction factors, and a check is performed to determine whether the amplitude value at the critical frequency is reduced compared to the current amplitude spectrum and / or whether the amplitude value reaches or falls below a predefined or stored threshold (e.g., an EMC requirement). Preferably, the correction factor is set so that the amplitude value at the critical frequency becomes minimal or at least falls below the threshold.

[0024] For example, the correction factor is one or two orders of magnitude smaller than the associated time parameter.

[0025] A single (rectangular) PWM pulse can, for example, be described by the following frequency spectrum:

[0026] Here, Fpuise is the frequency spectrum, co is the (circular) frequency, lAmp is the current level or current amplitude during a pulse (PWM amplitude), and corresponds to the phase current, Ton is the pulse duration, To is the pulse position and j is the imaginary unit ( -1 ).

[0027] A pulse train is the sum of individual pulses. This applies to both the frequency and time domains, so the frequency spectrum F(w) for n individual pulses is as follows:

[0028] Here, i is the running index over the sum of the n individual pulses. For equidistant pulses, for example, To(i) = i x Pulse spacing. If only one PWM pulse is generated in a pulse period, the pulse spacing is equal to the period duration (Tperiod). Each period can have a different period duration. Preferably, all phases or two phases per period have a PWM with different pulse widths and pulse heights.

[0029] The amplitude spectrum is the magnitude of the frequency spectrum |F(CÜ)|. The idea of ​​the invention is to replace one of the time parameters T with T + Tcorr, where Tcorr is the correction factor, and to vary the correction factor Tcorr such that the magnitude of the frequency spectrum at the critical frequency (|F(wcritical)|), i.e., the amplitude value, is minimized or at least reduced.

[0030] In one possible implementation, the pulse position (To), the pulse duration (Ton), or the period duration (Tperiode) is used as the time parameter. For example, the pulse duration Ton« is replaced by Ton« + T on - corr(i), where Tonkorr© is the correction factor for the pulse duration at run index i. When the pulse duration (Ton©) is used as a time parameter, it is accepted that a mean phase or motor voltage for the electric motor may be temporarily disturbed.

[0031] Another possibility is to vary the period duration (Tperiod) of the PWM in a suitable manner. If Tperiod is replaced by Tperiod + Tperiodekorr, where Tperiodekorr is the correction factor for the period duration, Tperiodekorr can be selected such that | F(cükritisch)| is minimal. Preferably, the period duration Ton is varied such that an average phase voltage before and after the change in the correction factor is essentially identical. In other words, the period duration Ton is adjusted accordingly to maintain a desired average phase voltage for the electric motor.

[0032] The following – without loss of generality – explains in more detail the variant in which the pulse position To© is varied. For this purpose, To© is replaced by To© + Tokorr©, where Tokorr© is the correction factor for the pulse position for the running index i. The corresponding frequency spectrum is thus

[0033] According to the procedure, Tokorr© is selected or adjusted in such a way that | F(ou>kntisch)| is minimal or at least reduced.

[0034] To reduce the computational effort, the method preferably considers all past and next signal pulses, i.e. the next one to be generated. This means that the sum of the frequency spectrum of the last signal pulse (n-1 ) and the frequency spectrum for the next signal pulse (n) is calculated, so that as an approximation The correction factor is preferably always calculated only for the next period consisting of up to three individual pulses.

[0035] In an advantageous further development, the correction factor is limited during the change by an upper limit (TkorrMax). In other words, a maximum value for the correction factor is limited so that |Tkorr(i)| < TkorrMax. Such a limitation ensures that a PWM control or PWM regulation for motor operation (e.g., using field-oriented control) is disturbed or influenced as little as possible. A quasi-continuous value range is therefore specified for the correction factor, and for correction factors within this value range, the calculated amplitude spectrum with the lowest amplitude value at the critical frequency is sought. For example, the upper limit or maximum value is dimensioned to approximately 1 ps (microsecond). Alternatively, the upper limit can be dimensioned to approximately 1% of the time parameter to be corrected, so that the correction factor can be varied, for example, within a range of ±1% of the time parameter.

[0036] In a particularly simple embodiment with reduced computational effort, a value range with a number of discrete correction values ​​is stored for the correction factor. Instead of a continuous value range, a set of stored correction values ​​or correction times is specified. Preferably, at least two, for example three, correction values ​​are stored as a value range, e.g. Tcorr(n) e {-1 / zs, 0 / zs, 1 / zs}. The amplitude spectrum is calculated for each correction value, with the correction value selected as the value for the correction factor for performing the PWM at which the calculated amplitude spectrum has the lowest or minimum amplitude value at the critical frequency. This ensures particularly simple implementation of the method. In one conceivable embodiment, the resulting amplitude spectrum is measured.In particular, the amplitude (kmp) of the amplitude spectrum at the critical frequency is measured or determined. With regard to the approximation for determining the frequency or amplitude spectrum described above, the amplitude for the frequency spectrum of the previous signal pulse (lAmp(ni)) is measurable. In particular, the frequency or amplitude spectrum of the previous signal pulse can be measured, and the value for the amplitude derived from this. For this purpose, the current ripple or EMC interference is measured, for example, by an average measurement over several milliseconds (ms) or by a peak measurement in the range of a few milliseconds and then converted into the frequency or amplitude spectrum using a Fourier transformation. During active use, preferably only the value for the critical frequency(ies) is calculated.

[0037] In a useful further development, the measured amplitude (lAmp(ni)) is used in the calculation of the frequency or amplitude spectrum, so that the amplitude spectrum can be predicted as accurately as possible.

[0038] The amplitude of the future current ripple (lAmp(n)) is unknown at the time of calculation. As an approximation, in the simplest case, the amplitude of the past signal pulse can be estimated as the value for the upcoming amplitude (lAmp(n) = lAmp(ni)). An estimate based on the current components Id, Iq of a field-oriented motor control with corresponding rotation is also possible. Such estimates are subject to a certain degree of error, but this can already be a sufficiently good approximation to reduce the amplitude at the critical frequency sufficiently to meet a specified EMC requirement.

[0039] Preferably, the amplitude of the future current ripple (lAmp(n)) is estimated when calculating the amplitude spectrum. "Estimate" or "estimate" here and below refers to an approximate determination of the future amplitude, in particular by evaluating the measured frequency spectrum for the past signal pulse, for example, by visual inspection, pre-characterized measurements, stored tables or characteristic curves, or by means of statistical-mathematical methods. During static motor operation, the phase voltages are approximately sinusoidal, so a comparatively simple estimation is possible. Alternatively, a motor model for the electric motor can be used to estimate or predict the amplitude.

[0040] In a preferred embodiment, the method is carried out for more than one critical frequency. This means that the amplitude values ​​at several critical frequencies are reduced. For example, the sum of the amplitude spectra for the critical frequencies is minimized, i.e. min(|F(cücriticali )| + | F(cücritical2)| ). Alternatively, a weighted sum can also be minimized, i.e. min(g1 x | F(oücriticali )| + g2*| F(ü0critical2)| ), where g1 and g2 are weighting factors. The weighting can, for example, be based on the respective amplitude values ​​at the critical frequencies, i.e. the frequency spectrum with the higher amplitude value at the critical frequency is weighted more heavily than the frequency spectrum with the lower amplitude value. Other cost functions are also conceivable.

[0041] The electric machine according to the invention is intended, and is suitable and configured, in particular as an electric motor drive in a motor vehicle. The electric machine comprises an electric motor, which is preferably designed as a brushless motor with a stator and a rotor rotatably mounted therein. The stator has a number of phase windings, which are connected, on the one hand, to the power converter and, on the other hand, are connected in a star connection, for example, at a common connection point (star point).

[0042] The electric machine further comprises a bridge circuit connected or coupled to the electric motor and a controller, i.e., a control unit. The bridge circuit is preferably part of a power converter, in particular an inverter. The controller is, for example, part of the power converter. The controller is generally suitable and configured—in terms of programming and / or circuitry—to carry out the method described above. The controller is thus specifically configured to add a correction factor to a time parameter of the PWM during operation and to adjust it such that an amplitude value of an amplitude spectrum is reduced at a critical frequency.

[0043] In a preferred embodiment, the controller is formed at least in its core by a microcontroller with a processor and a data memory in which the functionality for carrying out the method according to the invention is implemented in the form of operating software (firmware), so that the method - if necessary in interaction with a user - is carried out automatically when the operating software is executed in the microcontroller.

[0044] Within the scope of the invention, the controller can alternatively also be formed by a non-programmable electronic component, for example an ASIC (application-specific integrated circuit), in which the functionality for carrying out the method is implemented using circuitry.

[0045] The electric machine operated with the method thus exhibits improved performance with regard to EMC emissions and noise generated by the switching processes of the semiconductor switches. The method according to the invention is particularly suitable and configured for use in speed-controlled systems. However, its application is not limited to the automotive sector.

[0046] An additional or further aspect of the invention provides software on a medium or data carrier for carrying out or executing the method described above. This means that the software is stored on a data carrier and is intended for carrying out the method described above, as well as being suitable and configured for this purpose. This results in particularly suitable software for operating an electric motor, with which the functionality for carrying out the method according to the invention is implemented in programming terms. The software is thus, in particular, operating software (firmware), with the data carrier being, for example, a data memory of the controller.

[0047] An exemplary embodiment of the invention is explained in more detail below with reference to a drawing. The drawings show, in schematic and simplified representations:

[0048] Fig. 1 an electrical machine with a power source and an electric motor as well as with a power converter connected between them,

[0049] Fig. 2 three phase windings of a three-phase electric motor of the machine in star connection,

[0050] Fig. 3 a bridge module of a bridge circuit of the power converter for controlling a phase winding of the electric motor,

[0051] Fig. 4 an equivalent circuit diagram for the current source,

[0052] Fig. 5 is a block diagram for a pulse width modulation, and

[0053] Fig. 6 is a flowchart for a method for operating the electric motor.

[0054] The invention is explained below using a drive with a B6 circuit as an example. However, the invention can also be applied to other arrangements.

[0055] Corresponding parts and sizes are always provided with the same reference symbols in all figures.

[0056] Fig. 1 shows an electric machine 2 for an electric motor drive of a vehicle (not shown in detail), for example a motor vehicle or an electrically powered or drivable bicycle (e-bike). The machine 2 comprises a three-phase brushless electric motor 4, which is connected to a power source (voltage supply) 8 by means of a power converter (converter, inverter) 6. In this exemplary embodiment, the power source 8 comprises an internal vehicle energy storage device in the form of a (motor vehicle) battery 10, as well as a (DC voltage) intermediate circuit 12 connected thereto as part of an on-board electrical system, which extends at least partially into the power converter 6. The intermediate circuit 12 is essentially formed by a forward line 12a and a return line 12b, by means of which the power converter 6 is connected to the battery 10.The lines 12a and 12b are at least partially led into the power converter 6, in which an intermediate circuit capacitor 14 and a bridge circuit 16 are connected between them.

[0057] During operation of the machine 2, an input current IE supplied to the bridge circuit 16 is converted into a three-phase output current (motor current, three-phase current) lu, Iv, Iw for the three phases U, V, W of the electric motor 4. The output currents lu, Iv, Iw, hereinafter also referred to as phase currents, are fed to the corresponding phases (windings) U, V, W (Fig. 2) of a stator (not shown in detail).

[0058] Fig. 2 shows a star connection 18 of the three phase windings U, V, and W. The phase windings U, V, and W are each connected by a (phase) end 22, 24, 26 to a respective bridge module 20 (Fig. 3) of the bridge circuit 16, and are interconnected by the opposite end at a star point 28 as a common connection terminal. In the illustration in Fig. 2, the phase windings U, V, and W are each shown by means of an equivalent circuit in the form of an inductance 30 and an ohmic resistance 32, as well as a respective voltage drop 34, 36, 38.

[0059] The voltage drop 34, 36, 38 across the phase windings U, V, W is schematically represented by arrows and results from the sum of the voltage drops across the inductance 30 and the ohmic resistance 32 as well as the induced voltage 40. The voltage 40 (electromagnetic force, EMF) induced by a movement of a rotor of the electric motor 4 is represented in Fig. 2 by a circle.

[0060] The star connection 18 is controlled by the bridge circuit 16. The bridge circuit 16 is designed, in particular, as a B6 circuit with the bridge modules 20. In this embodiment, during operation, each of the phase windings U, V, W is switched at a high switching frequency between a high (DC) voltage level of the supply line 12a and a low voltage level of the return line 12b.

[0061] The high voltage level is, in particular, an intermediate circuit voltage UZK of the intermediate circuit 12, while the low voltage level is preferably a ground potential UG. This clocked control is implemented as a PWM control—illustrated by arrows in Fig. 1—by a controller 42, which enables control and / or regulation of the speed, power, and direction of rotation of the electric motor 4.

[0062] The bridge modules 20 each comprise two semiconductor switches 44 and 46, which are shown in Fig. 2 only schematically and as an example for phase W. On the one hand, the bridge module 20 is connected via a potential connection 48 to the supply line 12a and thus to the intermediate circuit voltage UZK. On the other hand, the bridge module 20 is connected via a second potential connection 50 to the return line 12b and thus to the ground potential UG. Via the semiconductor switches 44, 46, the respective phase end 22, 24, 26 of phases U, V, W can be connected either to the intermediate circuit voltage UZK or to the ground potential UG.

[0063] If semiconductor switch 44 is closed (conductive) and semiconductor switch 46 is opened (non-conductive, blocking), the phase ends 22, 24, 26 are connected to the intermediate circuit voltage UZK. Accordingly, when semiconductor switch 44 is opened and semiconductor switch 46 is closed, phases U, V, W are connected to ground potential UG. This makes it possible to apply two different voltage levels to each phase winding U, V, W using PWM control.

[0064] Figure 3 shows a simplified representation of a single bridge module 20. In this embodiment, semiconductor switches 44 and 46 are implemented as MOSFETs (metal-oxide semiconductor field-effect transistors), each of which switches between an on state and a off state in a clocked manner using PWM control. For this purpose, the respective gate terminals are connected to corresponding control voltage inputs 52, 54, via which the PWM control signals of controller 42 are transmitted.

[0065] Fig. 4 shows an equivalent circuit diagram for the power source 8. During operation, the battery 10 generates a battery voltage 11Bat and a corresponding battery current Ißat for operating the power converter 6. In Fig. 4, the internal resistance of the battery 10 is shown as an ohmic resistor 56 and a self-inductance of the battery 10 as an inductance 58. A shunt resistor 60 is connected in the return line 12b.

[0066] Depending on the switching states of the (power) semiconductor switches 44, 46, the phase currents Iu, Iv, Iw flow through the shunt resistor 60. The voltage drop across the shunt resistor 60 is amplified and evaluated. Using measurements and the knowledge of the switching states of the semiconductor switches 44, 46, the controller 42 reconstructs the phase currents Iu, Iv, Iw. Other measurement methods can also be used to determine the motor currents (e.g., direct phase current measurement). Together with the measured and / or calculated phase voltages (Ilu, Uv, Uw), the phase voltages (Uu, Uv, Uw) and the phase currents Iu, Iv, Iw are available to the controller 42.

[0067] In the embodiment of Fig. 1, the motor current is measured by an ammeter 62, for example, by means of the shunt resistor 60, and fed to the controller 42. The controller 42 controls and / or regulates motor operation based on motor variables, in particular based on the measured phase currents Iu, Iv, Iw and the calculated phase voltages Uu, Uv, Uw, as well as other variables (e.g., motor resistance, motor inductance, duty cycle of the PWM voltage). For example, a field-oriented control for the electric motor 4 is implemented here.

[0068] The switching operations of semiconductor switches 44, 46 during PWM generate pulsed currents in the intermediate circuit 12 or in the vehicle electrical system. These currents or current ripples can lead to electromagnetic compatibility (EMC) disturbances.

[0069] A method for operating the electric motor 4 is described below with reference to Fig. 5 and Fig. 6, which method is intended and configured to reduce such EMC interference.

[0070] Fig. 5 shows a schematic and simplified representation of two sections 66 and 68 arranged vertically one above the other, each showing a time diagram of the PWM (left) and an amplitude spectrum 64 (right) horizontally next to each other.

[0071] The timing diagrams, for example, show two PWM periods, each period being 50 ps long. Time t is plotted horizontally, i.e., along the x-axis or abscissa axis, while a PWM amplitude, for example, in volts, is plotted vertically, i.e., along a y-axis or ordinate axis (not shown in detail).

[0072] In the amplitude spectra 64, a (circular) frequency w is plotted along the X or abscissa axis, with the amplitude being plotted along a Y or ordinate axis (not shown in detail).

[0073] During pulse width modulation (PWM), for example, at least one signal pulse (PWM pulse) 70 is generated per period for each phase U, V, and W. The PWM is characterized by time parameters, in particular the period duration Tperiode, the pulse duration Ton, and the pulse position To. The pulse duration Ton essentially represents the turn-on time of the semiconductor switches 44, 46, i.e., how long the respective semiconductor switch 44, 46 is switched on. The pulse duration Ton is centered around the pulse position To within the period duration Tperiode.

[0074] According to the method, at least one of the time parameters is supplemented by a correction factor (correction time) Tkorr in order to effect a frequency shift of the amplitude spectrum 64 to EMC-non-critical frequencies through the modified PWM. Without limiting generality, an embodiment of the method is described below in which the pulse position To is replaced by a sum of the pulse position To and the correction factor Tkorr.

[0075] In a threshold comparison 72, the controller 42 first checks whether a current amplitude spectrum 64 of the EMC interference or current ripple meets a specified or stored EMC requirement. The EMC requirement is characterized, for example, by the fact that an amplitude threshold S must not be exceeded in the amplitude spectrum 64 in a specific frequency range.

[0076] The controller 42 thus checks in the threshold value comparison 72 whether the amplitude spectrum 64 reaches or exceeds the amplitude threshold value S. For this purpose, an average or peak measurement is carried out, for example, using the ammeter 62, and the amplitude spectrum 64 is determined from the current ripple using a Fourier transformation, in particular using a frequency spectrum, for example using an FFT (Fast Fourier Transform). If the amplitude threshold value S is reached or exceeded, the method is started. The corresponding frequency value of the amplitude spectrum 64 for the amplitude value A, at which the amplitude threshold value S is reached or exceeded, is referred to below as the critical frequency. Alternatively, the method can be carried out or started during the entire engine operation, regardless of the amplitude value A.

[0077] The critical frequency can be determined using special devices independently of the method according to the invention. For example, a resonance frequency of the electric machine 2 is determined.

[0078] Section 66 of Fig. 5 shows an exemplary operating situation in which the amplitude spectrum 64 has a frequency peak whose amplitude value A is greater than the amplitude threshold value S. In such an operating situation, the controller 42 starts the method described below. In a first method step 74, the pulse position To is replaced by the sum of the pulse position To and the correction factor Tcorr. Section 68 of Fig. 5 shows a corresponding PWM pulse sequence in which the next generated signal pulse 70 is shifted from the pulse position To by Tcorr in the period Tperiode.

[0079] In a subsequent second method step 76, an amplitude spectrum 64' is calculated, which is expected to result from the changed PWM time parameter (To + Tcorr) in the case of the EMC disturbance.

[0080] The amplitude spectra 64' is calculated as the magnitude of the frequency spectrum | F(oo)|. For the PWM pulse sequence, the frequency spectrum F(w) is approximately described by the following formula

[0081] The frequency spectrum F(w) is essentially composed of the frequency spectrum F n -i(w) of the previous pulse and the frequency spectrum F n (w) for the coming (future) pulse, where the frequency spectrum F n (w) is provided with the correction factor Tkorr. Preferably, the frequency component or frequency spectrum F(w) is calculated only for the critical frequency CUcritical.

[0082] The frequency spectrum F n-i(w) is in particular the measured amplitude spectrum 64 from the threshold comparison 72. For the frequency spectrum Fn(w), the amplitude lAmp is not known during the calculation and is estimated for the calculation, for example, using a motor model for the electric motor 4 and / or the intermediate circuit 12 (or on-board network). Alternatively, the amplitude lAmp obtained from the measured frequency spectrum F n -i(w) determined amplitude lAmp for the calculation of the frequency spectrum F n (w) is used. Method steps 74 and 76 are preferably repeated for a number of different correction factors Tcorr. Thus, several amplitude spectra 64' are calculated for different correction factors Tcorr. For this purpose, the value of the correction factor Tcorr is changed, and the corresponding amplitude spectrum 64' is subsequently calculated.

[0083] The correction factor Tkorr can be limited during change by an upper limit (Tkorr-Max). In other words, it is possible for a maximum value to limit the correction factor Tkorr so that |Tkonj < TkorrMax. Such a limitation ensures that a PWM control or PWM control for motor operation (e.g., using field-oriented control) is disturbed or influenced as little as possible. For example, the upper limit TkorrMax is dimensioned to approximately 1 ps (microsecond). Alternatively, the upper limit can be dimensioned to approximately 1% of the pulse position To, so that the correction factor Tkorr can be varied, for example, within a range of ±1% around the pulse position To.

[0084] Alternatively, a set or number of correction values ​​Tcorr can be stored. Preferably, at least two, for example, three, correction values ​​are stored as a value range, e.g., Tcorr e {-1 ps, 0 ps, ​​1 ps}.

[0085] In a subsequent method step 78, the calculated amplitude spectra 64' are compared with each other, and the correction factor Tkorr is selected at which the magnitude of the respective amplitude spectrum 64' at the critical frequency is quantically lowest. For a set MTkorr of different correction factors Tkorr, certainly.

[0086] In method step 80, the next signal pulse 70 is generated during PWM with the determined correction factor TKOR-. As shown, for example, in the amplitude spectrum 64' in section 68 of Fig. 5, no energy is removed from the amplitude spectrum 64 by the correction factor Tkorr; the spectrum, or rather the peak characterized by the critical frequency ÜJcritical and the amplitude value A, is frequency-shifted (at least partially) to a frequency cüneuw by the correction factor Tkorr.

[0087] The method described above can also be implemented for multiple critical frequencies. For example, the sum of the amplitude spectra for the critical frequencies is minimized. For two critical frequencies Wcriticali and Cücritical2, for example, in method step 78 certainly.

[0088] The claimed invention is not limited to the exemplary embodiments described above. Rather, other variants of the invention can also be derived therefrom by those skilled in the art within the scope of the disclosed claims without departing from the subject matter of the claimed invention. In particular, all individual features described in connection with the various exemplary embodiments can also be combined in other ways within the scope of the disclosed claims without departing from the subject matter of the claimed invention.

[0089] List of reference symbols for electrical machines

[0090] 4 electric motor

[0091] 6 power converters

[0092] 8 Power source

[0093] 10 Battery

[0094] 12 intermediate circuit

[0095] 12a Outward line

[0096] 12b Return line

[0097] 14 DC link capacitor

[0098] 16 bridge circuit

[0099] 18 star connection

[0100] 20 bridge modules

[0101] 22, 24, 26 phase end

[0102] 28 Star Point

[0103] 30 Inductance

[0104] 32 Resistance

[0105] 34, 36, 38 Voltage drop

[0106] 40 voltage

[0107] 42 controllers

[0108] 44, 46 semiconductor switches

[0109] 48, 50 Potential connection

[0110] 52, 54 control voltage inputs

[0111] 56 Resistance

[0112] 58 Inductance

[0113] 60 shunt resistance

[0114] 62 ammeters

[0115] 64, 64' amplitude spectrum

[0116] 66, 68 Section

[0117] 70 signal pulse

[0118] 72 Threshold comparison

[0119] 74, 76, 78, 80 Process step IE Input current

[0120] U, V, W phase lu, Iv, Iw phase current

[0121] UZK intermediate circuit voltage UG earth potential

[0122] Ußat battery voltage

[0123] I Bat Battery current s Amplitude threshold

[0124] A Amplitude value lAmp amplitude

[0125] CUcritical frequency

[0126] Tperiod period duration

[0127] Tone pulse duration

[0128] To pulse position Tcorr correction factor t time

[0129] (JL) Frequency

[0130] ÜJneu frequency

Claims

Claims 1. A method for operating an electric motor (4) by means of pulse width modulation, in which at least one pulse width modulated signal pulse (70) is generated at a pulse position (To) with a pulse duration (Ton) during a period (Tperiode), wherein a resulting amplitude spectrum (64) has an amplitude value (A) at a critical frequency (cücritical), - wherein at least one time parameter (Tperiode, To, Ton) of the pulse width modulation is replaced by a sum of the time parameter (Tperiode, To, Ton) and a correction factor (Tkorr), - wherein the correction factor (Tkorr) is changed such that a calculated amplitude spectrum (64') has a reduced amplitude value (kmp) at the critical frequency (cükritisch), and - where the pulse width modulation is carried out with the time parameter (Tperiode, To, Ton) and the changed correction factor (Tkorr).

2. Method according to claim 1, characterized in that the pulse position (To), the pulse duration (Ton) or the period duration (Tperiode) is used as the time parameter.

3. Method according to claim 2, characterized in that when the period duration (Tperiode) is used as a time parameter, the pulse duration (Ton) is varied such that an average phase voltage for the signal pulse (70) before and after the change in the correction factor (Tkorr) is substantially identical.

4. Method according to one of claims 1 to 3, characterized in that the correction factor (Tkorr) is limited during the change by an upper limit and / or a lower limit.

5. Method according to one of claims 1 to 4, characterized in that a value range with a number of discrete correction values ​​is stored for the correction factor (Tkorr), the amplitude spectrum (64') being calculated for each correction value, and that correction value is selected as the value for the correction factor (Tkorr) at which the calculated amplitude spectrum (64') has the minimum amplitude value (A) at the critical frequency (cücritical).

6. Method according to one of claims 1 to 5, characterized in that the amplitude spectrum (64) is measured.

7. The method according to claim 6, characterized in that an amplitude (kmp) for calculating a future amplitude spectrum (64') is determined from the measured amplitude spectrum (64).

8. Method according to one of claims 1 to 7, characterized in that an amplitude (kmp) is estimated during the calculation of the amplitude spectrum (64').

9. Method according to one of claims 1 to 8, characterized in that the method is carried out for more than one critical frequency (cücritical).

10. Electrical machine (2), comprising an electric motor (4) and a bridge circuit (16) as well as a bridge circuit (16) controlling Controller (42) for carrying out a method according to one of claims 1 to 9.

11. Software on a data carrier for carrying out a method according to one of claims 1 to 9, when the software runs on a computer.

Images