EP2810367A1

EP2810367A1 - Control of a permanent-magnet electric machine

Info

Publication number: EP2810367A1
Application number: EP13704181.0A
Authority: EP
Inventors: Sami ZAIM; Farid Meibody-Tabar; Babak NAHID-MOBARAKEH; Régis Bernard Albert MEURET
Original assignee: Universite de Lorraine; Hispano Suiza SA
Current assignee: Universite de Lorraine; Safran Electrical and Power SAS
Priority date: 2012-01-31
Filing date: 2013-01-21
Publication date: 2014-12-10
Also published as: US20140333244A1; CA2861954C; CA2861954A1; BR112014018174A2; WO2013114021A1; JP2015509357A; CN104094517B; CN104094517A; US9184682B2; FR2986389B1; BR112014018174B1; BR112014018174A8; FR2986389A1; RU2014135334A

Abstract

The invention relates to a method of controlling a permanent-magnet synchronous machine (3) comprising a stator and a rotor. The method comprises a step of determining an estimated position ($) of the rotor, a step of determining a second direct voltage setpoint (nudelta2*) which is alternately equal to a first direct voltage setpoint (nudelta1*) or equal to the first direct voltage setpoint (nudelta1*) plus a predetermined periodic signal (G). The step of determining an estimated position ($) of the rotor comprises a step of determining a coupling term (Deltaϊgamma), a step of determining a speed of rotation of the rotor (Omega5) as a function of said coupling term (Deltaϊgamma), and a step of determining the estimated position ($) of the rotor by integrating the speed of rotation of the rotor (Omega5).

Description

CONTROL OF AN ELECTRIC MACHINE WITH MAGNETS

PERMANENT

Background of the invention

The present invention relates to the control of a synchronous machine with permanent magnets.

It is known that to control a synchronous machine with permanent magnets, it is necessary to know at every moment the position of the rotor. Usually, position sensors are used to measure the position and calculate the speed of the machine. The main disadvantage of the use of these sensors (as well as processing cards that accompany them) is the decrease in the reliability of the system, which is a key element in the aeronautical field. The other disadvantages of this solution are the increase in weight, volume, and total cost of the system.

Many research works are therefore carried out in order to do without this position sensor, and thus to estimate the mechanical variables only from the measurement of the stator currents.

Several methods have already been proposed and validated to control a synchronous machine in medium and high speeds without position sensor. These methods are based on the estimation of the electromotive force vector (EMF) empty from the imposed voltages, the measurement of currents and equations of the machine. The EMF being directly proportional to the speed, the latter can also be estimated just like the position which is then obtained by simple integration of the speed. However, since the EMF is nil when stopped and embedded in measurement noise at low speeds, it is no longer observable in these areas of operation. Methods based on FEM estimation are therefore unsuitable for applications for which positional control is required.

To estimate the position at low speeds and at rest, the only remaining solution is to use the variations in the value of the stator inductors as a function of the position of the rotor. Several processes using inductance variations have already been proposed: According to a first type of method, the principle is to cut control every ten or twenty pulse width modulation (PWM) periods and to inject a high frequency signal (greater than the bandwidth of the current regulators). The ratio of the injected voltage to the variation of the measured current makes it possible to estimate the inductance, and since this depends on the position, the position can be estimated. An example is described in J. Kiel, A. Bunte, S. Beineke, "Sensor / ess torque control of permanent magnet synchronous machines over the whole operation rangé", EPEPEMC, TP-053, Dubrovnik & Cavat, September 2002.

According to a second type of method, the error on the estimation of the position is initially estimated. This error is regulated to zero using a corrector. The output of this corrector gives us an estimation of the speed, and by integration we get the estimated position of the rotor. In order to calculate the estimation error, the measurement of the currents just after the injection of the HF signal is compared with the current that should have been obtained theoretically if there was no RF signal.

The methods of the aforementioned first type have the following drawbacks: A direct calculation of the estimated position, it will therefore discontinuities each calculation. As the voltage references are calculated from the position of the rotor, the references will also know discontinuities which will cause torque surges that can be harmful.

The need to have to stop the order to make the estimate. Indeed, every ten or twenty MLI periods (depending on the desired accuracy), one of them is devoted solely to the injection of a high frequency signal for the estimation.

Over-twisting of the stator currents at the time of estimation is in some cases necessary.

The methods of the second type mentioned above have the following disadvantages:

In these methods, a measured current is compared to a current that would be obtained theoretically. So you have to have a accurate model of the engine if we want to ensure a good convergence of methods. These then become dependent on uncertainties as well as variations in the parameters of the machine.

- In addition, these methods do not apply to smooth rotor machines.

There is therefore a need to improve the control of a synchronous machine at low and zero speeds. Obiet and summary of the invention

The invention aims to meet this need by proposing a method of controlling a synchronous machine with permanent magnets comprising a stator and a rotor, said method comprising:

a step of determining an estimated position of the rotor,

a step of determining a direct current and a current in quadrature as a function of stator currents and the estimated position of the rotor,

a step of determining a first direct voltage setpoint and a quadrature voltage setpoint as a function of the direct current, the quadrature current, a direct current setpoint and a quadrature current setpoint,

characterized in that it comprises;

a step of determining a second direct voltage setpoint which is alternately equal to the first direct voltage setpoint or equal to the first forward voltage setpoint added with a predetermined periodic signal,

a step of determining stator voltage setpoints as a function of the second direct voltage set point, the quadrature voltage set point and the estimated rotor position,

a step of controlling said synchronous machine as a function of the stator voltage instructions,

wherein the step of determining an estimated rotor position comprises:

a step of determining a coupling term as a function of a difference between the current in quadrature when the second direct voltage setpoint is equal to the first direct voltage setpoint and the quadrature current when the second direct voltage setpoint is equal to the first direct voltage setpoint added with the predetermined periodic signal,

a step of determining a rotational speed of the rotor as a function of said coupling term, and

a step of determining the estimated position of the rotor by integrating the speed of rotation of the rotor.

Correlatively, the invention provides a control unit for permanent magnet synchronous machine comprising a stator and a rotor, said control unit comprising:

means for determining an estimated position of the rotor,

a module for determining a direct current and a current in quadrature as a function of stator currents and the estimated position of the rotor,

a module for determining a first direct voltage setpoint and a quadrature voltage setpoint as a function of the direct current, the quadrature current, a direct current setpoint and a quadrature current setpoint,

characterized in that it comprises:

a module for determining a second direct voltage setpoint which is alternately equal to the first direct voltage setpoint or equal to the first direct voltage setpoint added with a predetermined periodic signal,

a module for determining stator voltage setpoints as a function of the second direct voltage set point, the quadrature voltage set point and the estimated rotor position,

means for controlling said synchronous machine as a function of the stator voltage instructions,

wherein the means for determining an estimated position of the rotor comprises:

a module for determining a coupling term as a function of a difference between the quadrature current when the second direct voltage setpoint is equal to the first direct voltage setpoint and the quadrature current when the second direct voltage setpoint is equal to the first direct voltage setpoint added with the predetermined periodic signal,

a module for determining a rotational speed of the rotor as a function of said coupling term, and

a module for determining the estimated position of the rotor by integrating the speed of rotation of the rotor.

The step of determining a rotational speed of the rotor according to said coupling term may include the implementation of a corrector for canceling the term coupling.

Preferably, the predetermined periodic signal is a pulse signal.

According to one embodiment, the step of controlling said synchronous machine according to the stator voltage setpoints comprises supplying said stator voltage setpoints to a pulse width modulation inverter having a predetermined period, said second voltage setpoint direct current being equal to the first direct voltage setpoint added with the predetermined periodic signal for one to three periods of the pulse width modulation, every 15 to 25 periods.

The rotor may be a rotor with salient poles. The rotor may also be a smooth-pole rotor, said method comprising a step of saturation of stator teeth facing the poles of the rotor.

The invention also proposes a control system comprising a control unit according to the invention, an inverter and a synchronous machine.

The invention also relates to a computer program comprising instructions for performing the steps of a control method according to the invention when said program is executed by a computer.

This program can use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other form desirable shape. The invention also relates to a recording medium or information carrier readable by a computer, and comprising instructions of a computer program as mentioned above.

The recording media mentioned above can be any entity or device capable of storing the program. For example, the medium may comprise storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or a magnetic recording medium, for example a floppy disk or a disk. hard.

On the other hand, the recording media may correspond to a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The program according to the invention can be downloaded in particular on an Internet type network.

Alternatively, the recording media may correspond to an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question. Brief description of the drawings

Other features and advantages of the present invention will emerge from the description given below, with reference to the accompanying drawings which illustrate embodiments having no limiting character. In the figures:

FIG. 1 is a diagram of a control system according to one embodiment of the invention,

FIG. 2 represents reference marks relating to the actual and estimated positions of the rotor of the synchronous machine of the system of FIG. 1, and

FIG. 3 is a diagram of the control system of FIG. 1, in which the operation of the control unit is represented by functional modules.

Detailed description of embodiments

FIG. 1 represents a control system of a synchronous machine with permanent magnets according to one embodiment of the invention. The system of Figure 1 comprises a unit of command 1, a pulse width modulated inverter 2, and a permanent magnet synchronous machine 3.

The synchronous machine 3 comprises a rotor carrying permanent magnets and a stator having three-phase windings. Note Vabc three-phase statoric voltages and labc three-phase stator currents. The synchronous machine 3 can be characterized by different magnitudes, in particular by its dynamic statoric inductances. The synchronous machine 3 is a machine in which the dynamic stator inductors are dependent on the position of the rotor. It can therefore be a synchronous machine with protruding rotor, but also a synchronous machine with a smooth rotor whose stator teeth facing the rotor magnets are either slightly saturated or made slightly saturated by imposing a sufficiently high positive value of the rotor. stator current along the estimated direct axis. The structures of such machines are known to those skilled in the art and are therefore not described in detail.

The inverter 2 supplies the three-phase voltages Vabc for the synchronous machine 3 from a supply voltage (not shown), by modulation in pulse width, as a function of voltage setpoints Vabc * provided by the control unit. control 1. The operation of such an inverter 2 is known to those skilled in the art and is therefore not described in detail.

The control unit 1 determines the three-phase voltage setpoints Vabc * to be supplied to the inverter 2 to control the synchronous machine 3. For this purpose, the control unit 1 estimates the rotor position of the synchronous machine 3 by based on the variation of the stator dynamic inductances as a function of the position of the rotor, as explained below.

In the embodiment shown, the control unit 1 presents the hardware architecture of a computer and comprises a processor 4, a non-volatile memory 5, a volatile memory 6 and an input-output interface 7. processor 4 makes it possible to execute computer programs stored in the non-volatile memory 5 by using the volatile memory 6. The operation of the control unit 1 described hereinafter results from the execution of such a program. The input-output interface 7 makes it possible in particular to obtain the measurement of the currents of the synchronous machine 3 and supply the voltage setpoints Vabc * to the inverter 2.

In a variant, the control unit 1 is a digital control device of the DSP card, microcontroller or FPGA type,

With reference to FIG. 2, the principle used by the control unit 1 to estimate the rotor position of the synchronous machine 3 is now explained.

It is known that the voltages Vabc can be expressed by a direct voltage Vd and a voltage in quadrature v _q in a d, q reference linked to the rotor of the synchronous machine 3. Similarly, the currc currents can be expressed by a direct current i _d and a current in quadrature i _q in the d, q. FIG. 2 represents the d, q and an angle Θ which represents the position of the rotor with respect to a reference axis a.

Consider a protruding pole motor, the evolution of the inductance proper of a phase, as a function of the position Θ of the rotor neglecting the higher harmonics is in the following form:

L _S (0) = L _S + L _s .cos (20 + 0 _O ) (2)

This function consists of a constant part (the average value) and a variable part whose period is equal to 180 ° electrical, θο depends on the phase considered.

In the reference d, q, the equations of the synchronous machine 3 are (valid even in saturated mode):

d_ ^v d

-?, I ₂

dt

where Ρ (θ) is the rotation matrix defined by

case Θ sin Θ

Ρ (θ) = (A)

sin Θ cos Θ

(i _d , i _q ), (v _d , v _q ) and (if) _d , yj _q ) are the direct and quadrature components of the stator currents, voltages and total fluxes, Ω is the actual mechanical speed of the synchronous machine 3, R _s the resistance of the windings and p the number of pairs of poles. Considering only the first harmonic of the magnetomotive forces at the stator and neglecting the cross saturation effect between the fictitious windings d and q (m _dq = 0), we have:

(5)

where l _d and l _q are respectively the direct and quadrature dynamic inductances.

We limit ourselves here to the low speeds and to the stop, so we make the hypothesis that the terms proportional to the speed Ω are negligible. Equation (3) becomes simpler and becomes:

d ⁱ d

- ^v d ^{~ R} s'd (6)

dt

^• - v - R _s i _q

dt

We can notice that now the equations are completely decoupled. The injection of a voltage on the axis d will cause no response current on the axis q.

In the absence of a position sensor in the synchronous machine 3, the control unit 1 does not have access to the position Θ of the rotor and thus determines an estimated position 9. In FIG. δ, γ related to the estimated position § of the rotor. The estimation error between the two marks is represented by the angle φ (φ = 9 - Θ). The two marks rotate with respect to the stator at the electrical speed ρΩ for the real reference (Ω is the mechanical speed of the rotor), and ρΩ ₅ for the estimated reference (Ω ₅ is the estimated mechanical speed of the rotor).

By applying a rotation of φ using the matrix P (cp), the equation (3) in the reference δ, γ becomes; with

Again, as we are moving at low speeds, we can simplify these equations by neglecting the terms proportional to Ω:

After simplification, these equations take the form of ten

I d ^l q ^" = -R _s (l _q - (l _q -l _d ) sin ² φ) ί _δ + (l _g - (l _q -l _d ) * ί ^ηΔ φ) - ν _δ

dt

R 1

+ f pl _d l _q Q. _s + - † (l _q -l _d ) sin2 <p) - i _r - (l _q -l _d ) sin2q> - ν _γ (12)

di r _

d ^l q -R _s (l _d + (l _q ~ l _d ) sin <p) - i ₇ + (l _d + (l _q -l _d ) sin φ) ν _γ

dt

- (P ^l d ^l q ^a s ^l q -) sin 2 <p) - i _s -j (l _q -l _d ) sin 2 <p-v _s (13)

We can observe that at present, the equations of the synchronous machine 3 in the estimated δ, γ mark are again coupled.

This coupling depends at the same time on the saliency of the machine (more precisely the dynamic magnetic saliency, that is to say the difference between the direct and quadrature dynamic inductances, term (l _q - l _d )) and the estimation error φ.

In other words, in case of saliency and error on the estimation of the position, a variation of the component δ of the voltage will show a coupling term noted Δι _γ on the component γ of the stator current.

This term disappears if the error is canceled.

FIG. 3 shows the use of this principle in the control unit 1. In FIG. 3, the operation of the control unit 1 is represented in the form of functional modules which can correspond to the execution of a computer program by the processor

4 of Figure 1. The control unit 1 comprises a current regulator 10, a periodic signal generator 11, an adder module 12, a conversion module 13, a conversion module 14, a determination module 15, a speed estimator 16 and a integrator 17.

The control unit 1 works in the reference δ, γ estimated and handles in particular the following quantities:

a direct current setpoint ί * and a quadrature current setpoint i _Y *,

a forward voltage setpoint ν ι *, a forward voltage setpoint ν _δ2 * and a quadrature voltage setpoint v _Y *,

a periodic high frequency signal G,

an estimated speed Q _s of the rotor,

- estimated position $ of the rotor.

The current regulator 10 determines the direct voltage setpoint ν _δ ι * and the quadrature voltage setpoint v _Y * as a function of the direct current i ₅ , the quadrature current i _Y , the direct current setpoint ϊ _δ * and of the current setpoint in quadrature i _Y *. The realization of such a current regulator is known to those skilled in the art and is therefore not described in detail.

The periodic signal generator 11 provides the periodic signal G high frequency. By "high frequency" is meant a frequency lower than the frequency of the pulse width modulation of the inverter 2 but greater than the cutoff frequency of the current regulators. In this example, the periodic signal G is a voltage pulse signal. The amplitude of these pulses is chosen to be large enough to observe a significant coupling term in equation (13). However, this amplitude must not be too large either to the risk of disrupting the control and increase the losses in the synchronous machine 3. The skilled person is able to perform an appropriate dimensioning from these indications.

The adder module 12 determines the forward voltage setpoint ν _δ2 * as a function of the forward voltage setpoint v _5i * and of the periodic signal G. More precisely, the forward voltage setpoint ν _δ2 * is alternately equal to the forward voltage setpoint ν _δ ι * (ν _δ2 * = ν _δ ι *) or equal to the forward voltage setpoint ν _δ ι * added with a periodic signal G (ν _δ2 * = ν _δ ι * + G). For example, ν _δ2 * is normally equal at ν _δ ι * and, every 20 periods of the pulse width modulation of the inverter 2, ν _δ2 * is equal to ν _δ ι * + G during one to three periods of the pulse width modulation of the inverter 2.

The conversion module 13 converts the voltage readings of the reference mark δ, γ estimated at the values of the stator mark abc. In other words, the conversion module 13 determines the stator voltage setpoints Vabc * for the inverter 2, as a function of the forward voltage setpoint ν _δ2 *, the quadrature voltage setpoint v _Y * and the estimated position § of the rotor. The realization of such a conversion module is known to those skilled in the art and is therefore not described in detail.

The conversion module 14 converts the stator current Iabc measured in the synchronous machine 3 into current in the reference δ, y estimated. In other words, the conversion module 14 determines the direct current ΐ _δ and the quadrature current i _Y as a function of the stator currents Iabc and the estimated position 9 of the rotor. The realization of such a conversion module is known to those skilled in the art and is therefore not described in detail.

The determination module 17 determines a coupling term Δί _γ which appears on the component y of the current when the periodic signal G is injected on the component δ of the voltage. As explained above, the coupling term Δί _γ appears in case of estimation error on the position of the rotor and saliency. More specifically, the determination module 17 determines the coupling term Δϊ _γ by calculating the difference between the quadrature current i _Y when ν _δ2 * = ν _δ1 * just before the pulse and the current in quadrature i _Y when ν _δ2 * = ν _δ ι * + G. It is considered that during the impulse ν _δ1 * remains slowly variable.

The speed estimator 16 determines the estimated speed Ω of the rotor ₅ according to the _Δί γ coupling term. More precisely, it is known that the coupling term Δί _γ disappears if the error vanishes. The speed estimator 16 will therefore regulate the coupling Δϊ _γ to zero using a corrector. The output of this corrector will provide an estimate of the speed. Indeed, depending on the sign of the coupling, it is possible to know if the estimated mark is ahead or behind the actual mark. The corrector will therefore increase the estimated speed or, on the contrary, slow it down, in order to make the two marks coincide. The corrector used is for example a Corrector proportional-integral (PI), particularly interesting from the point of view of the calculation time. However, it is also possible to use other types of corrector. The PI corrector will be of the form:

Q _s = Q ₀ (Al _r + λ _r - dt) (14)

where Ω ₀ and T are the parameters of the estimator determining the convergence of the estimate and its dynamics.

The integrator 17 determines the estimated position 9 of the rotor by integrating the estimated speed Q _s :

In equations (12) and (13), the trigonometric terms, a function of the position error cp, have a period of n (180 ° electrical). Therefore, this error can converge either to 0 or to n depending on the initial position error. Indeed, if the initial error is too important (| <p |> n / 2), the method may converge to an error of n and the command without sensor can fail. Similarly, for a smooth machine, if the initial error is too large, it is more difficult to make the machine slightly saturated to obtain a sufficient gap between l _d and l _q . Indeed, it is the presence of the term (l _d - l _q ) in (12) and (13) which ensures the convergence of the estimator. If the initial error is not low enough, the ampere-turn imposed by the stator currents may not have a direct component sufficient to saturate the machine. In particular, in the case where the initial error is greater than n / 2, it is even possible to desaturate the machine and not to cause magnetic saliency. To address these problems and determine the initial estimated position 9 (0), an initialization method described in K. Tanaka, R. Moriyama, I. Miki, "Initial Rotor Position Estimation of Interior Permanent Synchronous Magnet" can be used. Motor Using Optimal Voltage Vector ", Electrical Engineering in Japan, vol. 156, no. 4, July 2006.

The estimated position 9- provided by the integrator 17 is used in particular by the conversion modules 13 and 14, as previously described.

The determination of the estimated position 9 made by the control unit 1 has several advantages. ; - The injection of voltage pulses is in the δ axis and not in the γ axis. When the error becomes weak, the axis δ merges with the axis d and thus the torque produced by the currents resulting from the pulses becomes negligible and does not disturb the control of the synchronous machine 3. Moreover, in this case, the component d of the stator current caused by these pulses participates in the saturation of the magnetic circuit and thus increases the saliency and facilitate convergence.

- A calculation of the position which is done by integration of an estimated speed and not in a direct way. There is no discontinuity on the estimate and the current references.

- The estimation process is simple and light and is therefore accompanied by a reduced calculation time.

- Robustness obtained, knowledge of the parameters of the synchronous machine 3 and their variations is not required.

The invention is particularly suitable for avionics applications such as flight control, braking system, landing gear output or any system using electric actuators equipped with synchronous permanent magnet motors, for which it is imperative to be able to make control in position and thus ensure a couple even when stopped.

In the embodiment described above, the periodic signal G is a pulse signal. In an alternative embodiment, the periodic signal G is a high frequency sinusoidal signal. The current response to the injected sinusoidal voltages then gives an estimate of the inductances Ι _δ , l _{Y as} well as the mutual m _5v (see relation (10)). The latter being the image of the estimation error <p, it can be corrected by the estimation module 16 to a zero value. This solution is more difficult to implement than that based on a periodic signal G composed of pulses. Indeed, unlike a pulse, it is not easy to inject using the inverter 2 (whose switching frequency is fixed by the pulse width modulation) a signal whose frequency must be much greater than the electrical frequency of the control signals so as not to disturb the regulation. In addition, in order to process current responses obtained, it will be necessary to use a bandpass filter centered on the frequency of the injected signal.

Claims

A method of controlling a permanent magnet synchronous machine (3) comprising a stator and a rotor, said method comprising;

a step of determining an estimated position ($) of the rotor,

a step of determining a direct current (ί _δ ) and a current in quadrature (i _Y ) as a function of stator currents (labc) and the estimated position (9 ·) of the rotor,

a step of determining a first direct voltage setpoint (ν _δ ι *) and a quadrature voltage setpoint (v _Y *) as a function of the direct current (i ₅ ), of the current in quadrature (i _Y ), a direct current setpoint (k *) and a quadrature current setpoint (i _Y *), characterized in that it comprises:

a step of determining a second direct voltage setpoint (V _Ô2 *) which is alternately equal to the first direct voltage setpoint (ν _δ1 *) or equal to the first forward voltage setpoint (ν _δ ι *) added to with a predetermined periodic signal (G),

a step of determining stator voltage setpoints (Vabc *) as a function of the second direct voltage setpoint (ν _δ2 *), the quadrature voltage setpoint (v _Y *) and the estimated position (9) of the rotor,

a step of controlling said synchronous machine (3) according to the stator voltage instructions (Vabc *),

wherein the step of determining an estimated position () of the rotor comprises:

a step of determining a coupling term (Δί _γ ) as a function of a difference between the quadrature current (i _Y ) when the second direct voltage set point (ν _δ2 *) is equal to the first voltage set point direct (ν _δ1 *) and the quadrature current (i _Y ) when the second direct voltage setpoint ( _δ 2 *) is equal to the first direct voltage setpoint (ν _δ1 *) added with the predetermined periodic signal (G) ,

a step of determining a rotational speed of the rotor (Ω ₅ ) as a function of said coupling term (Δί _γ ), and - A step of determining the estimated position (9 ·) of the rotor by integrating the speed of rotation of the rotor (Ω ₅ ).

2. The control method according to claim 1, wherein the step of determining a rotational speed of the rotor (Ω ₅ ) as a function of said coupling term (i _Y ) comprises the implementation of a corrector intended to cancel the coupling term (Δί _ν ).

3. Control method according to one of claims 1 and 2, wherein the predetermined periodic signal (G) is a pulse signal.

4. Control method according to one of claims 1 to 3, wherein the step of controlling said synchronous machine according to the stator voltage setpoints (Vabc *) comprises providing said stator voltage instructions (Vabc *) to a pulse width modulation inverter (2) having a predetermined period, said second forward voltage setpoint (v _{2 2} *) being equal to the first forward voltage setpoint (ν _δ1 *) added with the predetermined periodic signal ( G) for one to three periods of the pulse width modulation, every 15 to 25 periods.

5. Control method according to one of claims 1 to 4, wherein said rotor is a rotor with salient poles.

6. Control method according to one of claims 1 to 4, wherein said rotor is a rotor with smooth poles, said method comprising a step of saturation of stator teeth facing the poles of the rotor.

7. Computer program comprising instructions for implementing a control method according to one of claims 1 to 6 when said program is executed by a computer.

A computer readable recording medium comprising a computer program according to claim 7.

9. Control unit (1) for synchronous machine (3) with permanent magnets comprising a stator and a rotor, said control unit (1) comprising:

means for determining an estimated position ($) of the rotor,

a module (14) for determining a direct current (ί _δ ) and a current in quadrature (i _Y ) as a function of stator currents (labc) and the estimated position ($) of the rotor,

a module (10) determining a first direct voltage setpoint (ν _δ ι *) and a quadrature voltage setpoint (v _Y *) as a function of the direct current (ϊ _δ ), of the current in quadrature ( i _Y ), a direct current setpoint (ί _δ *) and a quadrature current setpoint (i _Y *),

characterized in that it comprises;

a module (12) for determining a second direct voltage setpoint (ν _δ 2 *) which is alternately equal to the first direct voltage setpoint (ν _δ ι *) or equal to the first direct voltage setpoint (ν _δ ι *) added with a predetermined periodic signal (G),

a module (13) for determining stator voltage setpoints (Vabc *) as a function of the second forward voltage setpoint (ν _δ2 *), the quadrature voltage setpoint (v _Y *) and the estimated position ( S) of the rotor,

means (7) for controlling said synchronous machine (3) as a function of the stator voltage instructions (Vabc *),

wherein the means for determining an estimated position ($) of the rotor comprises:

a module (15) for determining a coupling term (Δί _γ ) as a function of a difference between the quadrature current (i _Y ) when the second direct voltage set point (ν _δ2 *) is equal to the first direct voltage setpoint (ν _δ ι *) and the quadrature current (i _Y ) when the second direct voltage setpoint (ν _δ2 *) is equal to the first direct voltage setpoint (ν _δ ι *) added with the signal predetermined periodic period (G),

a module (16) for determining a rotational speed of the rotor (Ω ₅ ) as a function of said coupling term (Δί _γ ), and - A module (17) for determining the estimated position (9-) of the rotor by integrating the rotational speed of the rotor (<¾).

10. Control system comprising a control unit (1) according to claim 9, an inverter (2) and a synchronous machine (3).