System and method for controlling a powertrain reducing low-frequency current components connected to a control using pulse width modulation

Abstract

The invention relates to a method (100) for controlling a powertrain comprising a battery, an electric machine and an inverter connected at the input to the battery and at the output to the electric machine, the method comprising steps of: - determining (102) an electrical operating frequency of the electrical machine; - determining (114) a modulation frequency for applying duty cycles to switches of the inverter, wherein, at least for certain types of pulse width modulation and certain conversion rates, the determined modulation frequency is equal to the electrical frequency multiplied by a non-integer factor such that the product, expressed as an absolute value, of the difference between the non-integer factor and the integer closest to the non-integer factor, with the electrical frequency, is greater than a predetermined frequency threshold.

Description

[0001] DESCRIPTION

[0002] Title of the invention: System and method for controlling a powertrain reducing the low-frequency current components associated with a control system using pulse-width modulation

[0003] The present invention relates to the field of electrical engineering, and more specifically concerns a method and control system for a powertrain, finding particular application in the automotive industry.

[0004] An electric or hybrid vehicle includes a high-voltage battery, on the order of several hundred volts, an electric machine capable of supplying torque to the vehicle's wheels via a transmission chain, and an inverter capable of supplying alternating voltage to the terminals of the stator windings of the electric machine, from the direct voltage supplied by the high-voltage battery.

[0005] The inverter is conventionally controlled using pulse-width modulation. For example, the inverter might have three switching arms, each with an output point, a low switch connected to the output point and the negative terminal of the high-voltage battery, and an high switch connected to the output point and the positive terminal of the high-voltage battery. A distinct duty cycle is applied to each high switch over a modulation period, inducing three separate currents at the output of the three inverter output points. These three currents flow through the three phases of the electrical machine. The corresponding voltages of the electrical machine, across the stator windings, have an amplitude, frequency, and phase that can be adjusted according to the applied duty cycle values.It is recalled here that a duty cycle is the ratio between, on the one hand, the closing time of the switch to which this duty cycle is applied, and on the other hand, the modulation period.

[0006] The modulation period must be sufficiently small compared to the electrical period of the electric machine, that is, compared to the inverse of the electrical frequency of the voltages applied to the stator windings of the electric machine. This patent application uses pulse-width modulation (PWM) types with a high modulation frequency, typically on the order of 10 kHz (kilohertz) for a consumer automotive application. This electrical period can vary very frequently and very rapidly, as it depends on the engine speed. Typically, at the beginning of each modulation period, a control task is executed to determine the duty cycles to be applied to the inverter based on measurements of the stator currents of the electric machine, the high-voltage battery voltage, the position of the electric machine's rotor, and the rotor speed.

[0007] The electrical frequency of the electric machine can typically vary from 0 Hz (Hertz) at no motor speed to approximately 1000 Hz at maximum motor speed. The electrical period of the electric machine can therefore also vary from 1 ms (one millisecond) to a period greater than one second. For simplicity, the modulation period of the inverter is chosen to be constant across its entire operating range, for example, set at 100 µs (microseconds). In this case, pulse-width modulation is described as "asynchronous" because it is independent of the electrical frequency of the electric machine.

[0008] The invention is of course transposable to other frequency ranges corresponding to other types of rotating machines, for example a range of 0Hz to 200Hz or of 0Hz to 2000Hz.

[0009] The use of asynchronous pulse-width modulation means that the supply voltages to the electrical machine are never perfectly periodic according to the machine's electrical frequency, which is determined by the motor speed, since the modulation frequency is not a multiple of this electrical frequency. The true period of the supply voltages is, in fact, the lowest common multiple of the modulation period and the electrical period determined by the motor speed.

[0010] Subharmonics with frequencies lower than the electrical frequency of the electric machine therefore exist in the case of asynchronous pulse-width modulation. Figure 1 illustrates an example of this effect of asynchronous pulse-width modulation. The black bars represent the amplitudes in volts (V) of voltage harmonics h, in the case where the electrical frequency f eThe machine's frequency is 768 Hz and the modulation frequency is 10 kHz (kilohertz). The voltage amplitude at the electrical frequency f e is represented by a grey bar.

[0011] We observe the presence of a subharmonic with a frequency fb equal to 16Hz. Indeed, we have: 16Hz = 10kHz - 13*768Hz. The presence of such a subharmonic poses problems when its absolute frequency approaches 0Hz, when the voltage amplitude of this subharmonic is relatively large, and when the frequency impedance of the electrical machine at this subharmonic frequency is low.

[0012] Indeed, these conditions imply that this low-frequency subharmonic creates a non-negligible low-frequency current component on each power supply phase of the electrical machine.

[0013] Therefore, at constant torque, the electrical machine must be able to withstand more current in its stator, incurring greater electrical losses and a decrease in efficiency, compared to a situation where these conditions on subharmonics are not present. Furthermore, this low-frequency current component causes the stator to heat up more quickly and can cause a maximum current threshold to be exceeded in one of the stator phases. The instantaneous current in this phase is at least equal to the sum of the current at the electrical frequency and this low-frequency current component, which can reach several tens of amperes. This overcurrent can trigger a hardware protection mechanism by immediately opening the switches, which are generally transistor-based and include circuits to protect against excessive current or temperature.

[0014] In other words, this low-frequency current component can cause degraded operation of the electrical machine during some of its operating points.

[0015] On the other hand, this low-frequency component can travel on the vehicle's high-voltage bus and disrupt the operation of other high-voltage components. To overcome these problems, synchronous pulse-width modulation (PWM) is used, meaning that the modulation frequency is chosen as a multiple of the electrical frequency of the electric machine. Specifically, in pre-calculated synchronous PWM, for each operating point of the machine, assuming a steady-state operation, Fourier analysis allows us to predict precisely the harmonic content of the electric machine's voltages, based on an average modulation frequency over a pre-selected electrical rotation period. This frequency determines the number of switching instants for the switches, which are the decision variables to be optimized.The demodulation frequency is then chosen so as to avoid voltage harmonics of large amplitudes.

[0016] This type of strategy is, however, poorly suited to automotive applications, where the operating point varies rapidly over large ranges of values. In particular, during strong vehicle acceleration, pulse-width modulation will never be truly synchronous, because each new modulation frequency is calculated only at the beginning of each modulation period, and not continuously.

[0017] Furthermore, it is difficult to accurately determine the engine speed, and therefore the electrical frequency, of the electric machine due to the vibrations and disturbances inherent in the vehicle's mechanical operation. Indeed, these vibrations and disturbances prevent the engine speed from being truly constant over an electrical period.

[0018] On the other hand, measuring the electrical frequency uses an electrical rotor position sensor, whose electrical signals are themselves subject to disturbances and therefore require filtering. Synchronous modulation thus necessitates an end-to-end analysis of how the engine speed measurement is performed, and signal processing to eliminate the disturbances and vibrations to which this measurement is subject. This requires expensive resources and also increases the complexity of the powertrain control.

[0019] Finally, the inventors found that to completely avoid any asynchronism problems, not only must the modulation frequency be a multiple of the machine's electrical frequency, but it is also preferable for the electrical voltages to be in phase with the modulation periods, in order to reduce the harmonic content of the inverter's output voltage. This latter condition implies a very precise instantaneous measurement of the rotor's position, even under significant acceleration, and therefore a costly position sensor.

[0020] The present invention aims to remedy, at least in part, the aforementioned drawbacks by providing a method and a control system for a powertrain that prevents the occurrence, during operation, of low-frequency current components on the order of tens of amperes, due to subharmonics, while avoiding the use of expensive sensors or complex engine speed measurement algorithms requiring significant computing resources and resulting in increased hardware costs. It should be noted that in this application, low frequencies are defined as frequencies close to 0 Hz relative to the electrical frequency of the electric machine, for example, frequencies below 20% of this electrical frequency. This notion of low frequency can be reduced to a frequency threshold for which the frequency impedance of the electric machine is sufficiently high. For example, the order of magnitude for a 1000 Hz machine is around 200 Hz.

[0021] To this end, the invention proposes a method for controlling a powertrain comprising at least one battery, an electric machine, and an inverter connected at the input to the battery and at the output to the electric machine.

[0022] The ordering process includes the following steps:

[0023] - determination of an electrical frequency f e of the operation of the electrical machine,

[0024] - selection of a pulse-width modulation type to drive the inverter, - calculation of control voltages to be applied to the electrical machine,

[0025] - Calculation of the inverter conversion rate based on the amplitude of the calculated control voltages,

[0026] - Determination of a modulation frequency for applying duty cycles to inverter switches,

[0027] the control method being characterized in that, at least when the selected pulse width modulation type is continuous modulation and when the inverter conversion rate is strictly greater than a first conversion rate threshold, or when the selected pulse width modulation type is discontinuous modulation and when the inverter conversion rate is strictly greater than a second conversion rate threshold, then the determined modulation frequency is equal to the electrical frequency multiplied by a non-integer factor such that the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the nearest integer to the non-integer factor, with the electrical frequency, is greater in absolute value than a predetermined frequency threshold.

[0028] It is understood that the step of determining the modulation frequency fulfills the condition imposed by the invention on the product of the difference between on the one hand the non-integer factor and on the other hand the integer closest to the non-integer factor, with the electrical frequency, whatever the electrical frequency determined, provided that the type of pulse width modulation selected is a continuous modulation with an inverter conversion rate strictly greater than the first conversion rate threshold, or a discontinuous modulation with an inverter conversion rate strictly greater than the second threshold.

[0029] As is known, for a three-phase electrical machine, the inverter conversion ratio is the ratio between, on the one hand, the amplitude of the fundamental of the three-phase voltage controlled at the input of the stator of the electrical machine, and on the other hand, the amplitude of the DC voltage applied to the input of the inverter multiplied by a factor equal to 1 / √3. This factor must be adapted in the case of an electrical machine with more than three phases.

[0030] The first threshold is, for example, equal to 0.9 and the second threshold is, for example, equal to 0.8. The first threshold can, of course, be equal to the second threshold. Other values ​​for the first and second thresholds are also possible; they are, for example, set according to the specifications of the powertrain. The predetermined frequency threshold ensures that the absolute frequency of a voltage subharmonic due to the asynchronous nature of the pulse-width modulation used is not too close to zero, and therefore that there is no low-frequency current component with a significant amplitude in a continuous modulation situation with a conversion factor close to, or greater than, one (i.e., in overmodulation), or in a discontinuous modulation situation with a sufficiently high conversion factor. These situations generate voltage harmonics of significant amplitudes.The predetermined frequency threshold is, of course, strictly greater than zero; for example, it is between 50 and 200 Hz. It depends on the maximum amplitude of the acceptable low-frequency current harmonic according to the specifications of the vehicle to which the invention applies. For illustrative purposes only, on an 110 kW electric motor powered by a 400 V battery, the predetermined frequency threshold is, for example, 150 Hz.

[0031] The predetermined frequency threshold is, for example, determined so that the corresponding frequency impedance of the electrical machine is greater than a predetermined impedance threshold. In other words, at the frequency of a voltage subharmonic, since this frequency is greater in absolute value than the predetermined frequency threshold, the impedance of the electrical machine is sufficiently high to prevent the amplitude of the corresponding current subharmonic from being excessive. The predetermined impedance threshold is, for example, approximately 200 mΩ (milliohms), or more commonly between 150 and 300 mΩ.

[0032] In an alternative embodiment, the predetermined frequency threshold is determined such that the amplitudes of low-frequency current harmonics induced by the application of duty cycles are below a predetermined current amplitude threshold. This predetermined current threshold is, for example, between 2 and 15 amperes and, for example, set at 10 amperes.

[0033] Thanks to this invention, the constraints associated with synchronous pulse-width modulation are overcome by using asynchronous pulse-width modulation in which the modulation frequency is chosen so that the aforementioned condition regarding the ratio between the modulation frequency and the machine's electrical frequency is met. This condition can be easily verified and fulfilled at the beginning of each modulation period, even when the electrical frequency is determined with little precision or varies slightly. Therefore, the modulation frequency does not necessarily need to be determined at the beginning of each modulation period, but can be determined, for example, every two or three modulation periods.

[0034] The electrical operating frequency of the electric machine is, in this application, as explained in relation to the prior art, the inverse of the fundamental period of the voltages applied to the stator of the electric machine. This electrical frequency is determined by measuring the rotor speed. As is known, it is given the value of the motor speed in revolutions per minute, multiplied by the number of pole pairs of the electric machine's rotor, and divided by sixty, in the case of a synchronous machine. In the case of an asynchronous machine, the determination of the electrical frequency also takes into account the slip of the electric machine.

[0035] The modulation frequency is the inverse of a modulation period, the inverter switch undergoing a maximum of two state changes during a modulation period, while it is driven in pulse width modulation.

[0036] According to an advantageous and optional feature of the invention, when the selected pulse-width modulation (PWM) type is continuous modulation and the inverter conversion rate is below the first conversion rate threshold, or when the selected pulse-width modulation (PWM) type is discontinuous modulation and the inverter conversion rate is below the second conversion rate threshold, then the modulation frequency is set to a predetermined frequency. This feature simplifies inverter control by maintaining a constant modulation frequency throughout the operation of a pulse-width modulation that does not generate high-amplitude, low-frequency current subharmonics.This predetermined modulation frequency can be the same regardless of the electrical frequency of the electrical machine if it is sufficiently high, for example, greater than or equal to 10 kHz. The predetermined modulation frequency is chosen, for example, to be more than ten times the electrical frequency. Indeed, using such a predetermined frequency is sufficient, in cases where the inverter's conversion rate is quite low, to avoid problematic low-frequency voltage harmonics.

[0037] In an alternative embodiment, when the selected pulse width modulation type is continuous modulation and the inverter conversion rate is below the first conversion rate threshold, or when the selected pulse width modulation type is discontinuous modulation and the inverter conversion rate is below the second conversion rate threshold, the modulation frequency is calculated based on the operating point of the electrical machine.

[0038] In yet another embodiment, the determined modulation frequency is equal to the electrical frequency multiplied by a non-integer factor such that the product of the difference between the non-integer factor and the nearest integer to the non-integer factor, with the electrical frequency, is greater in absolute value than the predetermined frequency threshold, regardless of the type of asynchronous pulse-width modulation used and regardless of the inverter conversion ratio.

[0039] In yet another embodiment of the invention, when the selected pulse width modulation type is continuous modulation and when the inverter conversion rate is greater than a first conversion rate threshold, or when the selected pulse width modulation type is discontinuous modulation and when the inverter conversion rate is greater than a second conversion rate threshold, then the determined modulation frequency is equal to the electrical frequency multiplied by a non-integer factor such that the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the nearest integer to the non-integer factor, with the electrical frequency, is greater in absolute value than a predetermined frequency threshold.In this embodiment, when the selected pulse width modulation type is continuous modulation and when the inverter conversion rate is strictly less than the first conversion rate threshold, or when the selected pulse width modulation type is discontinuous modulation and when the inverter conversion rate is strictly less than the second conversion rate threshold, the modulation frequency is, for example, equal to a predetermined frequency, or calculated based on the operating point of the electrical machine.

[0040] In other words, strict inequality conditions on the first and second thresholds can be replaced by non-strict inequality conditions, and vice versa, without changing the nature of the invention.

[0041] The invention also relates to a powertrain control system comprising at least one battery, an electric machine and an inverter connected at the input to the battery and at the output to the electric machine,

[0042] the control system comprising means for:

[0043] - Determining the operating frequency of the electrical machine, - Selecting a pulse-width modulation type to drive the inverter, - Calculating the control voltages to be applied to the electrical machine.

[0044] - Calculation of the inverter conversion rate based on the amplitude of the control voltages,

[0045] - Determination of a modulation frequency for applying duty cycles to inverter switches,

[0046] the control system being characterized in that, at least when the type of pulse width modulation selected by the selection means is continuous modulation and when the inverter conversion rate is strictly greater than a first conversion rate threshold, or when the type of pulse width modulation selected by the selection means is discontinuous modulation and when the inverter conversion rate is strictly greater than a second conversion rate threshold, the modulation frequency determination means are configured to determine the modulation frequency as being equal to the electrical frequency multiplied by a non-integer factor, such that the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the integer closest to the non-integer factor, with the electrical frequency, is greater in absolute value than a predetermined frequency threshold.

[0047] The control system according to the invention implements the control method according to the invention.

[0048] In the control system according to the invention, the predetermined frequency threshold is, for example, such that the corresponding frequency impedance of the electrical machine is greater than a predetermined impedance threshold.

[0049] In an alternative embodiment, in the control system according to the invention, the predetermined frequency threshold is determined so that the amplitudes of low frequency current harmonics induced by the application of duty cycles are less than a predetermined current amplitude threshold.

[0050] According to an optional feature of the control system according to the invention, when the selected pulse-width modulation type is continuous modulation and the inverter conversion rate is below the first conversion rate threshold, or when the selected pulse-width modulation type is discontinuous modulation and the inverter conversion rate is below the second conversion rate threshold, the modulation frequency determination means are configured to select a modulation frequency equal to a predetermined frequency. This predetermined modulation frequency is, for example, greater than ten times the electrical frequency.

[0051] The powertrain control system according to the invention has advantages similar to those of the powertrain control method according to the invention.

[0052] Other features and advantages of the invention will become apparent from the following description on the one hand, and from several illustrative and non-limiting examples of embodiments given with reference to the attached schematic drawings on the other hand, in which:

[0053] [fig 1] already commented on in relation to the prior art, is a spectral representation of the voltage harmonic amplitudes of an electrical machine powered by an inverter driven by asynchronous pulse-width modulation, corresponding to a particular electrical frequency and modulation frequency,

[0054] [Fig. 2] represents an electric powertrain and a control system according to the invention of the electric powertrain, in one embodiment of the invention, and

[0055] [Fig. 3] represents steps of a control method according to the invention, for the electric powertrain of Figure 2, in one embodiment of the invention. According to one embodiment of the invention shown in Figure 2, a powertrain of a vehicle incorporating a control system 40 according to the invention, comprises a high-voltage battery 28, with a voltage Vbatt between its terminals of several hundred volts.

[0056] An inverter 24 is connected in parallel to the battery 28 via a smoothing capacitor 26 and battery relays 32, 34. The inverter 24 supplies a three-phase electric machine 22, capable of operating as a motor to propel the vehicle or for energy recovery during braking. In this embodiment, the powertrain is entirely electric and the electric machine 22 is three-phase, but in alternative embodiments the powertrain may include a combustion engine in addition to an electric machine, and the electric machine may have more than three power phases.

[0057] The three-phase 22 electric machine is represented here only by three stator windings L1, L2, L3, mounted in a star configuration in this embodiment and therefore connected together to a neutral point N.

[0058] The 24 inverter has three switching arms, each comprising:

[0059] - an exit point respectively M1, M2, M3,

[0060] - a low switch respectively 1_L, 2_L, 3_L connected on one side to the respective output point M1, M2, M3 and on the other side to the negative terminal of battery 28 via the battery relay 34, and

[0061] - a high switch respectively 1_H, 2_H, 3_H connected on one side to the respective output point M1, M2, M3 and on the other side to the positive terminal of the battery 28 via the battery relay 32.

[0062] The upper switches 1_H, 2_H, 3_H and lower switches 1_L, 2_L, 3_L are controlled switches, for example, transistors. The control of the upper switches 1_H, 2_H, 3_H and the lower switches 1_L, 2_L, 3_L uses a control circuit, included in the control system 40, which may also include a vehicle computer. The control system 40 implements, in software, the control method 100, which will be described later in relation to Figure 3.

[0063] The output points M1, M2, M3 are connected to the free ends of the stator windings respectively L1, L2, L3, opposite the neutral point N.

[0064] During vehicle operation, the currents in the stator windings L1, L2, and L3 (and possibly in the rotor when wound), the battery voltage Vbatt 28, the position of the electric machine's rotor 22, and the rotor's rotational speed are measured. From these measurements and a setpoint for the electric machine's operating point, an algorithm implemented by the control system 40 determines new voltages to be applied to the electric machine 22. These new voltages are expressed as duty cycles a1, a2, a3, to be applied to the upper switches 1_H, 2_H, and 3_H respectively, while the lower switches 1_L, 2_L, and 3_L apply duty cycles 1-a1, 1-a2, and 1-a3 respectively, neglecting dead time.

[0065] Each duty cycle is applied over a modulation period, meaning that the high switches 1_H, 2_H, and 3_H will each be closed at most once during a modulation period, and respectively a1%, a2%, and a3% of the time during the modulation period. In other words, each high switch undergoes at most two state changes during the modulation period. The inverse of the modulation period is the modulation frequency, denoted f. m in this request.

[0066] Applying these duty cycles generates a current II in the stator winding L1, a current 12 in the stator winding L2, and a current 13 in the stator winding L3, with an electrical frequency denoted f e in this request. This electrical frequency f e is also the electrical frequency of the voltages across the stator windings L1, L2, L3. The electrical frequency f eis proportional to the rotor speed, except for slip in the case of an asynchronous machine.

[0067] The powertrain control method 100 is now described in relation to Figure 3. This control method 100 is executed, for example, at each modulation period, regardless of whether the machine's operating point has changed. The control method 100 is implemented in a vehicle computer, for example, a dedicated powertrain control computer within the control system 40, or it is implemented in the vehicle's main computer.

[0068] The control method 100 includes a step 102 for determining an electrical frequency f e operating frequency of the electrical machine, corresponding to the motor speed at the calculated operating point. This electrical frequency f ealso corresponds to the electrical frequency of the control voltages that must be applied to the electrical machine 22.

[0069] This step of determining 102 an electrical frequency f e The operation of the electrical machine is followed by a selection step 104 of a pulse-width modulation type to drive the inverter 24. In this embodiment of the invention, in this step 104, either low-speed spatial vector modulation or high-speed FTB (Flat Top Bottom) modulation is selected to limit electrical losses by reducing switching. Spatial vector modulation is continuous modulation, while FTB modulation is discontinuous modulation. The modulations described here are examples.

[0070] Then, the control method 100 calculates the control voltages to be applied to the stator windings L1, L2, L3 of the electric machine 22, based on, among other things, an operating point of the electric machine 22, measurements of the stator currents in the electric machine 22, the voltage Vbatt of the high-voltage battery 28, the position of the rotor of the electric machine 22, and the rotor speed. The operating point is calculated by a process other than control method 100, implemented by the computer less frequently than control method 100, for example, every 10 milliseconds. During this other process, an operating point of the machine, corresponding to the current setpoints to be obtained in the stator windings L1, L2, L3, is determined based on a setpoint motor torque, a measurement of the motor speed, and the voltage Vbatt at the input of the inverter 24.

[0071] The calculation step 106 of the control voltages is followed by a calculation step 108 of a conversion rate of the inverter, equal to the ratio between on the one hand the amplitude of the fundamental of the calculated control voltages, and on the other hand the amplitude of the DC voltage applied to the input of the inverter multiplied by a factor equal to 1 / √3.

[0072] Depending on the type of pulse-width modulation selected, and the calculated conversion rate, the control method 100 determines 110 how it should determine 112 or 114 the modulation frequency f m , for the application of duty cycles to the inverter 24 allowing the calculated control voltages to be applied to the stator windings L1, L2, L3.

[0073] When the selected pulse-width modulation type is continuous modulation (therefore, in this case, spatial vector modulation), and when the calculated conversion rate is less than a first conversion rate threshold, for example, set at 0.9, or when the selected pulse-width modulation type is discontinuous modulation (therefore, in this case, FTB modulation) and when the calculated conversion rate is less than a second conversion rate threshold, for example, set at 0.8 (N-branch), then the modulation frequency f m is determined to be equal to a predetermined frequency, for example, 10 kHz. The modulation frequency f m The predetermined frequency is possibly also chosen to be greater than ten times the electrical frequency f eto ensure that any low-frequency voltage subharmonic has a very small amplitude. For example, the predetermined modulation frequency is chosen from several predetermined modulation frequencies depending on the electrical frequency f e The predetermined frequency can, for example, be a function of the machine's operating point, for example, be set at 2kHz below 500rpm (revolutions per minute), at 5kHz below 8000rpm or at 10kHz below 15000rpm, with, for example, a hysteresis effect depending on this operating point or depending on the duration of application of this predetermined modulation frequency.

[0074] In an alternative embodiment, when the selected pulse width modulation type is continuous modulation and when the calculated conversion rate is less than a first conversion rate threshold, or when the selected pulse width modulation type is discontinuous modulation and when the calculated conversion rate is less than a second conversion rate threshold, the modulation frequency can be freely chosen, for example based on criteria such as the temperature of the power electronics.

[0075] When the selected pulse-width modulation type is continuous modulation (therefore, in this case, spatial vector modulation), and when the calculated conversion rate is strictly greater than the first conversion rate threshold, or when the selected pulse-width modulation type is discontinuous modulation (for example, among others, FTB modulation) and when the calculated conversion rate is strictly greater than the second conversion rate threshold (Y branch), then the modulation frequency f m is determined 114 so that, the modulation frequency f m being equal to the electrical frequency f e multiplied by a non-integer factor mf, the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the integer closest to the non-integer factor, with the electrical frequency f e , is greater in absolute value than a predetermined frequency threshold fmm.

[0076] For example, in this determination step 114, the control system 40 tests a modulation frequency f m , for example, equal to a predetermined frequency of 10 kHz or any other value. We have:

[0077] m f = f m / f e = ⌊m f ⌋ + {m f} ⇒ f m = ⌊m f ⌋f e + {m f}f e

[0078]

[0079] I

[0080] The non-integer factor m f is decomposed into an integer part ⌊m f ⌋ and a decimal part {m f}.

[0081] ⌊m f ⌋f e is a frequency synchronous with the electrical frequency, and

[0082]

[0083] is a frequency asynchronous with the electrical frequency f e .

[0084] In this determination step 110, if the asynchronous frequency

[0085]

[0086] and the asynchronous frequency (1

[0087]

[0088] — are greater in absolute value than the predetermined frequency threshold fmm, then the modulation frequency is the modulation frequency f m tested, otherwise the control system 40 repeats the test performed with a modulation frequency f m equal to another predetermined frequency, for example 11kHz, and so on until the asynchronous frequency

[0089]

[0090] and the asynchronous frequency (1 — {my] e ) of the new tested modulation frequency are greater in absolute value than the predetermined threshold f mm frequency. This predetermined frequency threshold (fmm) is, for example, between 50 and 200Hz.

[0091] Indeed, thanks to the modulation frequency f mdetermined, the voltage harmonics of the electrical machine 22 have frequencies f such that:

[0092] f = m fm + n fe = + n fe +, with MEN,;îl EZ,'The low frequency voltage harmonics corresponding to m strictly greater than one, have negligible amplitudes not inducing significant currents even with a low frequency impedance of the electrical machine 22 at these low frequencies.

[0093] The other low-frequency harmonics approaching 0Hz are therefore of frequencies

[0094]

[0095] and (i - that is to say greater in absolute value than the predetermined threshold fmm of frequency.

[0096] The predetermined frequency threshold fmm is determined such that the corresponding frequency impedance Z(fmm) of the electrical machine is greater than a predetermined impedance threshold. In other words, we have:

[0097] Z(fmin ) = √(R 2 + (2πLf min ) 2 ) > Z min

[0098] where R is the resistive component of the impedance of the electrical machine 22 and L its inductive component.

[0099] Z min is, for example, between 100 and 300 mΩ, which ensures that a low-frequency voltage subharmonic due to the choice of the modulation frequency f m , this voltage subharmonic being by construction of higher frequency in absolute value than the predetermined threshold fmm of frequency, has an amplitude not generating a low frequency current component of the order of one or several tens of amperes.

[0100] Of course, the invention is not limited to the examples just described, and many modifications can be made to these examples without departing from the scope of the invention. In particular, the features of different embodiments of the invention can be combined to carry out the invention, provided that these embodiments are not incompatible with each other.

Claims

DEMANDS 1- Control method (100) of a powertrain comprising at least one battery (28), an electric machine (22) and an inverter (24) connected at the input to the battery (28) and at the output to the electric machine (22), the control process (100) comprising the steps of: - determination (102) of an electrical frequency (f e ) of operation of the electrical machine (22), - selection (104) of a pulse width modulation type to drive the inverter (24), - calculation (106) of control voltages to be applied to the electrical machine (22), - calculation (108) of a conversion rate of the inverter (24) as a function of an amplitude of the control voltages, - determination (114) of a modulation frequency (f m ) for applying duty cycles to inverter switches (24), the control method (100) being characterized in that, at least when the selected pulse-width modulation type is continuous modulation and when the inverter conversion rate (24) is strictly greater than a first conversion rate threshold, or when the selected pulse-width modulation type is discontinuous modulation and the inverter conversion rate (24) is strictly greater than a second conversion rate threshold, then the modulation frequency (f m ) determined is equal to the electrical frequency (f e ) multiplied by a non-integer factor such that the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the integer closest to the non-integer factor, with the electrical frequency (f e ), is greater in absolute value than a predetermined frequency threshold. 2- Control method (100) of a powertrain according to claim 1, wherein the predetermined frequency threshold is determined so that the corresponding frequency impedance of the electric machine is greater than a predetermined impedance threshold.

3. A method of controlling (100) a powertrain according to claim 1, wherein the predetermined frequency threshold is determined such that the amplitudes of low-frequency current harmonics induced by the application of duty cycles are less than a predetermined current amplitude threshold.

4. A method of controlling (100) a powertrain according to any one of claims 1 to 3, wherein when the selected pulse-width modulation type is continuous modulation and when the inverter conversion rate (24) is less than the first conversion rate threshold, or when the selected pulse-width modulation type is discontinuous modulation and when the inverter conversion rate (24) is less than the second conversion rate threshold, the modulation frequency (f m ) is chosen equal to a predetermined frequency. 5- Method for controlling (100) a powertrain according to claim 4, wherein the predetermined modulation frequency (f m ) is chosen to be greater than ten times the electrical frequency (f e ). 6- Control system (40) of a powertrain comprising at least one battery (28), an electric machine (22) and an inverter (24) connected in input to the battery (28) and in output to the electric machine (22), the control system (40) comprising means for: - determination of an electrical frequency (f e ) of operation of the electrical machine (22), - selection of a pulse-width modulation type to drive the inverter (24), - calculation of control voltages to be applied to the electrical machine (22), - calculation of an inverter conversion rate (24) as a function of the amplitude of the control voltages, - determination of a modulation frequency (f m ) for applying duty cycles to inverter switches (24), the control system (40) being characterized in that, at least when the type of pulse-width modulation selected by the selection means is continuous modulation and when the conversion rate of the inverter (24) is strictly greater than a first conversion rate threshold, or when the type of pulse-width modulation selected by the selection means is discontinuous modulation and when the conversion rate of the inverter (24) is strictly greater than a second conversion rate threshold, the modulation frequency determination means (f m ) are configured to determine this as being equal to the electrical frequency (f e) multiplied by a non-integer factor, such that the product of the difference between, on the one hand, the non-integer factor and, on the other hand, the nearest integer to the non-integer factor, with the electrical frequency (f e ), is greater in absolute value than a predetermined frequency threshold. 7- Control system (40) of a powertrain according to claim 6, wherein the predetermined frequency threshold is such that the corresponding frequency impedance of the electric machine (22) is greater than a predetermined impedance threshold. 8- Control system (40) of a powertrain according to claim 6, wherein the predetermined frequency threshold is determined so that the amplitudes of low frequency current harmonics induced by the application of duty cycles are less than a predetermined current amplitude threshold. 9- A powertrain control system (40) according to any one of claims 6 to 8, wherein when the selected pulse-width modulation type is continuous modulation and when the inverter conversion rate (24) is below the first conversion rate threshold, or when the selected pulse-width modulation type is discontinuous modulation and when the inverter conversion rate (24) is below the second conversion rate threshold, the modulation frequency determination means (f m ) are configured to choose the modulation frequency (f m ) equal to a predetermined frequency. 10- Control system (40) of a powertrain according to claim 9, wherein the predetermined frequency is greater than ten times the electrical frequency (f e ).

Images