阅读说明：本技术 永磁同步电动机的启动方法和飞机压气机 (Starting method of permanent magnet synchronous motor and aircraft compressor ) 是由 温塞斯拉斯·布尔斯 帕斯卡·雅克·弗雷德里克·盖伊·杜泰恩 于 2019-09-24 设计创作，主要内容包括：本发明公开了一种永磁同步电动机的启动方法,所述永磁同步电动机包括转子和定子,所述定子包括分别与多个相接通的绕组,转换电路将所述多个相与电源接通以便控制所述同步电动机的转子旋转,所述转换电路包括向DC-AC转换器供电的DC-DC转换器,所述DC-AC转换器包括多个用于在多个连续的控制阶段控制所述转子的旋转的可控开关,所述方法包括：根据驱动表控制所述DC-AC转换器的开关以便确定所述电动机转子的加速度,所述驱动表为每个控制阶段与所述开关的状态对应表。根据预定的加速度ACC确定电角度A(E1-E2),以及如果电角度A大于预定的阈值角度A-(seuil),则发送控制改变信号Q(E3)。(The invention discloses a starting method of a permanent magnet synchronous motor, the permanent magnet synchronous motor comprises a rotor and a stator, the stator comprises a plurality of windings communicated with a plurality of phases respectively, a conversion circuit is used for communicating the plurality of phases with a power supply so as to control the rotation of the rotor of the synchronous motor, the conversion circuit comprises a DC-DC converter for supplying power to a DC-AC converter, the DC-AC converter comprises a plurality of controllable switches for controlling the rotation of the rotor in a plurality of continuous control phases, and the method comprises the following steps: controlling the switching of the DC-AC converter in order to determine the acceleration of the motor rotor according to a drive table, which corresponds to the state of the switching for each control phase. Determining the electrical angle A from the predetermined acceleration ACC (E1-E2), and if the electrical angle A is greater than the predetermined acceleration ACCThreshold angle A seuil Then a control change signal Q is sent (E3).)

1. A method of starting a permanent-magnet synchronous motor (100) comprising a rotor and a stator, the stator comprising windings respectively switched on with a plurality of phases (P1-P3), a switching circuit (400) switching the plurality of phases (P1-P3) on with a power supply (2) for controlling the rotation of the rotor of the synchronous motor (100), the switching circuit (400) comprising a DC-DC converter (410) supplying a DC-AC converter (420) comprising a plurality of controllable switches (T1-T6) for controlling the rotation of the rotor according to a plurality of successive control phases (S1-S6), the method comprising:

controlling switches (T1-T6) of the DC-AC converter (420) according to a drive table for each control phase (S1-S6) corresponding to the state of the switches (T1-T6) in order to determine the acceleration of the rotor of the electric motor (100),

the method is characterized by comprising the following steps:

determining an electrical angle A from a preset acceleration ACC (E1-E2), the determination of the electrical angle A comprising the steps of:

first step (E1): integrating said preset acceleration ACC to obtain a speed V according to the formula:

V(n)＝V(n-1)+T_S*ACC+K*V(n-1)

wherein:

v (n) is the instantaneous rotor speed at point n,

v (n-1) is the instantaneous rotor speed at the point in time n-1,

T_Sis a constant of sampling time that is,

ACC is a preset acceleration, K is a velocity correction constant;

second step (E2): integrating the speed V to derive the electrical angle A;

step (E3): if the electrical angle A is greater than a preset threshold angle A_seuilThe transmission control stage (S1-S6) changes the signal Q.

2. Method for starting a permanent-magnet synchronous motor (100) according to claim 1, characterized in that said drive table comprises six control phases (S1-S6), said preset threshold angle a_seuilIs 60 degrees.

3. A method of starting a permanent magnet synchronous motor (100) according to claim 1 or 2, characterized in that the method comprises: -detecting whether the rotor of the electric motor (100) is decelerating, and-if so-the value of the speed correction constant K is not equal to 0.

4. A method of starting a permanent magnet synchronous motor (100) according to any of claims 1-3, characterized in that the value of the speed correction constant K is 2-8%, preferably around 5%, if a deceleration of the rotor is detected.

5. A method for starting a permanent magnet synchronous motor (100) according to claim 3 or 4, characterized in that when the current I of the DC-AC converter (420) is_DCACWhen negative, it is determined that rotor deceleration is detected.

6. Method for starting a permanent-magnet synchronous motor (100) according to claim 3 or 4, characterized in that when the voltage U of the DC-AC converter (420) is high_DCACAs time increases, it is determined that rotor deceleration is detected.

7. Compressor (1) for an aircraft, characterized in that it comprises:

a permanent magnet synchronous motor (100) comprising a rotor and a stator, the stator comprising windings respectively in communication with a plurality of phases (P1-P3), a switching circuit (400) communicating the plurality of phases (P1-P3) with a power supply (2) for controlling the rotation of the rotor of the synchronous motor (100), the switching circuit (400) comprising a DC-DC converter (410) supplying a DC-AC converter (420) comprising a plurality of controllable switches (T1-T6) for controlling the rotation of the rotor in a plurality of successive control phases (S1-S6); and

calculator (500) configured to implement the method of starting a permanent magnet synchronous motor according to any of claims 1-6.

Technical Field

The invention relates to a method for starting a compressor, in particular for supplying fuel cells installed on board an aircraft with oxygen.

Background

Electrochemical reactions between different fluids in a fuel cell can produce electrical energy. Electrical energy is generated by supplying hydrogen and oxygen to such fuel cells and reacting them therein. Generally, the compressor supplies air to the fuel cell to supply oxygen to the fuel cell.

As shown in fig. 1, the compressor 1 comprises a permanent-magnet synchronous motor 100, the speed of the permanent-magnet synchronous motor 100 being controlled by a conversion circuit 200 of the "pulse amplitude modulation" type. As is known, the electric motor 100 comprises a rotor and a stator comprising windings able to generate a magnetic field according to a received current. In this example, the motor 100 includes three phases P1, P2, P3, each controlled by a converter circuit 200. The phases P1, P2, P3 turn on the stator windings of the motor 100. In order to control the motor 100 to operate at high speed, e.g., about 170,000rpm, the inverter circuit 200 must control the phases P1, P2, P3 of the motor 100 to a high frequency mode, e.g., about 5.5 KHz.

The above-described conversion circuit 200 includes a DC-DC converter 210 and a DC-AC converter 220 connected in series. The DC-DC converter 210 is connected to the power source 2 as an input terminal, and is connected to the DC-AC converter 220 as an output terminal to supply a voltage lower than the power source 2 to the DC-AC converter 220. The DC-AC converter 220 (also referred to as an inverter) includes a plurality of switches T1-T6, which are controlled to supply a desired current to each phase P1, P2, P3 of the motor 100. In a known manner, as shown in fig. 1, the DC-DC converter 210 and the DC-AC converter 220 are both controlled by the drive calculator 300 so as to accurately control the phases P1, P2, P3 of the motor 100.

To facilitate control of the rotation of the rotor of the motor 100, the current in the stator windings should be precisely controlled according to the position of the rotor relative to the stator. In order to determine the position of the rotor, the prior art has provided several solutions.

According to a first solution, patent WO2017/178752a1 discloses the installation of three hall sensors at the stator of the motor, which are circumferentially spaced at an angle of 120 ° in order to directly detect the position of the rotor. However, such sensors are expensive and cumbersome to install. In fact, it is desirable to detect the position of the rotor without adding equipment to the motor.

According to a second solution, it is disclosed in patent FR3028112a1 to determine the position of the rotor by observing the electromotive force values of the motor without the addition of sensors. However, this solution can only be implemented when the motor is in a steady state. Furthermore, this solution does not allow to control the starting, i.e. the transition phase, of the motor. In fact, the observation of the electromotive force value is only related to the minimum rotation speed of the motor.

In practice, to effect start-up, the drive calculator 300 includes the drive table shown in FIG. 2, which defines a plurality of successive control phases S1-S6. Each control phase S1-S6 determines the current control of the phases P1, P2, P3, i.e. the state of the switches T1-T6 (OFF 0 or ON 1 state), in order to apply the preset speed setpoint V1-V6 for a determined period D1-D6. In other words, the drive table defines a preset acceleration, represented in fig. 3 by the linear slope of the speed set-point Vc, to start the electric motor 100. However, this solution has drawbacks.

In fact, the driving table is theoretical and does not take into account the actual operating conditions of the motor 100, in particular the tightening torque of the bearings of the motor 100, which varies as a function of the following factors: the speed of the motor, the inductance of the converter that varies the peak current, the value of the voltage delivered by the power supply 2 to the compressor 1, the operating temperature of the compressor 1, the manufacturing tolerances of the motor 100, etc. In fact, the rotor speed of the motor 100 does not follow a linear acceleration, and there is a fluctuation in the actual acceleration, which produces a fluctuation in the linear slope of the speed VR with respect to the speed set value Vc, as shown in fig. 3, the speed V varying with time t. This fluctuation causes the motor 100 to repeatedly decelerate/accelerate, which increases motor wear and may prevent the motor from starting.

To optimize the start-up of the motor 100 to the maximum, the drive calculator 300 may increase the control setting of the DC-DC converter 210, but this may increase the wear on the electronic components. Furthermore, the acceleration can also be reduced in order to optimize the starting, but this entails a loss of starting speed of the motor.

A device for starting a brushless motor is also known from patent application WO2009/016939a 1. In particular, it discloses adjusting the speed by changing the phase of the electrical angle. In particular, it teaches to use static data to correct the value and timing of the electrical phase. In this patent, the static correction data is obtained from theoretical curves relating to the electrical, mechanical and environmental characteristics (motor and charging) of the entire system. The system does not allow for dynamic calibration. On the other hand, patent application WO2009/016939a1 does not teach the use of a DC-DC converter.

The present invention therefore aims to overcome these drawbacks by providing a new system and a new method for starting an aircraft permanent magnet synchronous motor, to achieve a robust and fast start of the motor without adding equipment to the motor. Another object is to limit velocity oscillations.

Although the invention was originally developed to address the fuel cell compressor problem, the invention is applicable to any start-up of a permanent magnet synchronous motor.

Disclosure of Invention

To this end, the invention relates to a method of starting a permanent-magnet synchronous motor comprising a rotor and a stator, the stator comprising windings communicating with a plurality of phases, respectively, a conversion circuit communicating the phases with a power supply for controlling the rotation of the rotor of the synchronous motor, the conversion circuit comprising a DC-DC converter supplying power to a DC-AC converter comprising a plurality of controllable switches for controlling the rotation of the rotor in a plurality of successive control phases, the method comprising: controlling the switching of the DC-AC converter in order to determine the acceleration of the motor rotor according to a drive table, which corresponds to the state of the switching for each control phase.

The method is remarkable in that the method comprises the following steps:

determining an electrical angle A according to a preset acceleration ACC; and

if the electrical angle A is greater than a preset threshold angle A_seuilThe control phase change signal Q is sent.

By means of the method of the invention, the control phase change is determined dynamically on the basis of the estimated motor speed, rather than from a preset table. This achieves an optimization of the motor acceleration to limit the starting time of the motor.

Preferably, the drive table comprises six control phases, the preset threshold angle being 60 °. Thus, the threshold angle corresponds to one sixth of a revolution of the motor.

Advantageously, the step of determining the electrical angle a comprises: a first step of integrating the preset acceleration ACC to derive a speed V and a second step of integrating the speed V to derive the electrical angle a.

Preferably, the first simple integration step is determined according to the following equation:

V(n)＝V(n-1)+T_s*ACC+W

wherein:

v (n) is the instantaneous rotor speed at point n,

v (n-1) is the instantaneous rotor speed at the point of time n-1,

-T_sis a constant of sampling time that is,

ACC is a preset acceleration, and

-W is a variable parameter.

Still preferably, the variable parameter W is determined according to the following formula:

W＝K*V(n-1)

where K is the velocity correction constant.

Thus, the current velocity V (n) is automatically corrected based on the previous velocity V (n-1).

Preferably, the method comprises an automatic step of detecting whether the rotor of the motor decelerates, the value of the speed correction constant K being unequal to 0 if deceleration of the rotor is detected. Preferably, the speed correction constant K has a value of 2% -8%, preferably around 5%, if a rotor deceleration is detected. This achieves that the deceleration is compensated and thus limited by controlling the electric motor. The speed set point can thus be adaptively adjusted.

According to an aspect of the invention, when the current I of the DC-AC converter is low_DCACWhen negative, it is determined that rotor deceleration is detected.

According to another aspect of the invention, when the voltage U of the DC-AC converter is high_DCACAs time increases, it is determined that rotor deceleration is detected.

More preferably, the step of detecting the deceleration of the rotor comprises the sub-step of detecting an increase in phase current of the motor over the control current. This makes it possible to easily detect deceleration.

The invention further relates to a compressor for an aircraft, comprising: a permanent magnet synchronous motor including a rotor and a stator, the stator including windings in respective communication with a plurality of phases, a switching circuit to switch the plurality of phases to a power source to control rotation of the rotor of the synchronous motor, the switching circuit including a DC-DC converter to supply power to a DC-AC converter, the DC-AC converter including a plurality of controllable switches to control rotation of the rotor in a plurality of successive control phases; and a calculator configured to implement the method as described above.

Preferably, the compressor is a fuel cell air compressor.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments are briefly introduced below.

FIG. 1 is a schematic diagram of a switching circuit and motor of a compressor of the prior art (described previously);

fig. 2 is a driving representation of a DC-AC converter of the conversion circuit of fig. 1;

FIG. 3 is a motor for the compressor of FIG. 1Velocity set value Vc and velocity V_RA schematic diagram of a curve of (a);

FIG. 4 is a schematic diagram of a switching circuit and motor of the compressor according to the present invention;

fig. 5 is a driving representation of the DC-AC converter of the conversion circuit of fig. 4;

fig. 6 is an electrical angle diagram of the DC-AC converter of fig. 4 at different control stages;

fig. 7 is a schematic diagram of the steps implemented by the calculator to transmit the phase change signal Q;

fig. 8 is a schematic view of the step of changing the signal Q by the speed correction transmission phase;

FIG. 9 is a first exemplary schematic diagram for determining a correction constant from a voltage measurement of a DC-AC converter;

FIG. 10 is a second exemplary schematic diagram for determining a correction constant from a current measurement of a DC-AC converter;

FIG. 11 is a schematic view of one embodiment of a motor starting method according to the present invention, an

FIG. 12 is an estimate V of the speed of the motor_EAnd a corrected speed set point Vc.

It should be noted that the appended drawings illustrate the invention in a detailed manner so as to implement the invention, and of course, may, of course, be used to better define the invention, if necessary.

Detailed Description

The invention will be explained in relation to an aircraft compressor motor. It goes without saying, however, that the present invention is applicable to any permanent magnet synchronous motor.

Fig. 4 schematically shows an embodiment of a compressor 1 according to the invention. In this example, the compressor 1 is a fuel cell compressor (not shown). Such a compressor 1 is configured to generate electric energy by providing oxygen-containing air to perform an electrochemical reaction. However, the invention is applicable to any type of aircraft compressor.

Still referring to fig. 4, the compressor 1 is powered by a power supply 2. The power supply 2 is configured to supply electrical power to the compressor 1. The compressor comprises an electric motor 100, a conversion circuit 400 connecting said electric motor 100 to a power supply 2, and a calculator 500 for controlling the conversion circuit 400.

The motor 100 is a permanent magnet synchronous motor including a rotor (not shown) and a stator (not shown). The rotor is cylindrical and the stator extends around the cylinder. The rotor has permanent magnets distributed around the cylindrical periphery. The stator has windings spaced circumferentially around the rotor. Preferably, the stator comprises three windings distributed 120 ° apart from each other. The switching circuit 400 is configured to supply power to the windings such that they generate an electromagnetic field that reacts with the permanent magnets in the rotor to cause it to rotate. Since the operation of such permanent magnet synchronous motors is known, it is not described in detail here. As shown in fig. 4, the motor 100 includes three windings. The motor 100 is a three-phase motor and is powered by three-phase (P1, P2, P3) current, with each phase P1, P2, P3 powering one winding, respectively.

The switching circuit 400 directs the current supplied by the power source 2 into the different windings of the motor 100 so as to continuously supply the latter with current, thereby effecting rotation of the rotor. To this end, the conversion circuit 400 includes a DC-DC converter 410 and a DC-AC converter 420.

The DC-DC converter 410 is configured to be supplied with a so-called "input" voltage by the power supply 2. The DC-DC converter 410 is configured to deliver a so-called "output" voltage to the DC-AC converter 420, which has a value different from the value of the input voltage. The DC-DC converter 410 is a boost type (also referred to as a "boost" converter) if the output voltage is higher than the input voltage, and is a buck type (also referred to as a "buck" converter) if the output voltage is lower than the input voltage. In the present embodiment, the DC-DC converter 410 is a step-down converter.

As shown in fig. 4, the DC-DC converter 410 includes a winding 411 configured to be charged by an input voltage. When winding 411 discharges, it increases the value of the output voltage. Since the operation of such a DC-DC converter 410 is known, it will not be described in detail here. To control the DC-DC converter 410, the calculator 500 is configured to control the maximum current value flowing through the winding 411 when the winding 411 is charged. This enables the winding 411 to be charged to obtain a desired output voltage value.

The DC-AC converter 420 (also referred to as an inverter) is configured to convert direct current to alternating current to power the various phases P1-P3 of the electric motor 100. The DC-AC converter 420 is configured to receive the direct current of the DC-DC converter 410, which is supplied by the power supply 2. The DC-AC converter 420 is configured to provide current to each of the phases P1-P3 of the motor 100.

As shown in fig. 4, the DC-AC converter 420 includes a plurality of controllable switches T1-T6 to direct current through the windings. As shown in FIG. 5, the states of the plurality of switches T1-T6 are defined according to the control phases S1-S6 in which they are located. In other words, the state of each of the switches T1-T6 is preset for each control phase S1-S6. This enables the power supply of each phase P1-P3 of the electric motor 100 to be determined during the control phases S1-S6. Switches T1-T6 are controlled by calculator 500 as described below.

The calculator 500 is configured to control the DC-DC converter 410 on the one hand and the DC-AC converter 420 on the other hand. As previously described, the calculator 500 is configured to send the current set value to the DC-DC converter 410. The current set point determines the maximum current flowing through winding 411 when charging the winding to determine the desired output voltage.

The calculator 500 is also electrically connected to the DC-AC converter 420 for control thereof. In particular, the calculator 500 controls the states of the switches T1-T6 according to the control phases S1-S6 from the table shown in FIG. 5. In this table, a value "0" corresponds to the off state of the switch, and a value "1" corresponds to the on state.

According to the invention, the calculator 500 is configured to vary the control phases S1-S6 according to the value of the electrical angle A. Thus, the changes of the control phases S1-S6 are not performed statically by reading predetermined time periods from the table, but are done dynamically in conjunction with the characteristics of the electric motor 100.

Fig. 7 shows a first embodiment of a method for starting the electric motor 100.

In this example, the calculator 500 includes: a sampler 501 configured to sample a preset acceleration ACC stored in the memory according to a sampling constant Ts; and an adder 502 that allows the previous speed V (n-1) to be added to determine the current speed V (n). Then, the preset acceleration ACC is integrated to determine the speed. The value of the acceleration ACC is defined to meet the required start-up time.

Mathematically, this first stage can be described by the following formula.

V(n)＝V(n-1)+T_s*ACC

Wherein:

-V (n) is the instantaneous rotor speed at point n,

v (n-1) is the instantaneous rotor speed at the point of time n-1,

-T_sis a constant of sampling time that is,

ACC is a preset acceleration.

Still referring to fig. 7, calculator 500 includes a saturator 503 placed at the output of adder 502 to limit the current speed v (n). The current velocity V (n) is transferred to memory 504, which provides the previous velocity V (n-1) to adder 502. During this first phase, the preset acceleration ACC is integrated to obtain the current speed v (n).

Similarly, calculator 500 implements integration of the current speed v (n) to obtain an electrical angle a (n) corresponding to the angular position of the rotor.

Also, still referring to fig. 7, the current velocity v (n) is transmitted to a sampler 505 configured to sample the current velocity v (n) according to a preset sampling constant. The calculator 500 further comprises an adder 506 enabling to add the previous electrical angle a (n-1) to determine the current electrical angle a (n). Similar to the previous, the calculator 500 comprises a saturator 507 placed at the output of the adder 506, so as to limit the value of the current electrical angle a (n).

The current electrical angle a (n) is transmitted to a memory 508 which provides the previous angle a (n-1) to the adder 506 and on the other hand the electrical angle a (n) is transmitted to a comparator 509, the comparator 509 being configured to compare the current angle a (n) with a threshold angle a_seuilA comparison is made. When the current angle A (n) is greater than the threshold angle A_seuilThe calculator 500 sends a phase change signal Q to change the switching states of the control phases S1-S6 and the switches T1-T6. In this example, calculator 500 includesA counter 510 to count the phase change signal Q to determine the current control phases S1-S6.

FIG. 6 has six control phases S1-S6, threshold angle A_seuilEqual to 60 deg., one sixth of a revolution. During a phase change Q, the value of the current electrical angle a will reset to zero in order to detect the threshold angle of the next phase, as shown in fig. 7.

Referring to fig. 6, the change in the current electrical angle a with time is represented in electrical degrees. Once the threshold angle is exceeded, the electrical angle A is reset to zero and the switches T1-T6 are switched to the next control phases S1-S6. This phase change is optimal because it is a function of the inherent characteristics of the motor 100. By means of the invention, the angular position of the rotor is accurately determined without the need for additional power tools.

According to a preferred aspect of the invention, the calculator 500 comprises a correction constant K in order to limit the speed fluctuations of the electric motor 100 around the speed set point determined by the control phases S1-S6. To this end, with reference to fig. 8, the calculator 500 comprises an adder 511 which allows to add a variable parameter W which is a function of the correction constant K and the speed V (n-1).

Corrected, the current speed v (n) is defined according to the following formula:

V(n)＝V(n-1)+T_s*ACC+W

wherein:

-W is a variable parameter determined according to the following formula:

W＝K*V(n-1)

where K is the velocity correction constant.

Several modes of calculation of the correction constant K are set forth next. The purpose is to detect whether a deceleration of the speed of the electric motor 100 occurs when the electric motor 100 is to be returned to the speed set point determined by the control phases S1-S6. By using the correction constant K, the speed set value can be increased, so that deceleration can be restricted.

According to the first embodiment, the voltage U in the DC-AC converter 420 is detected_DCACAn increase occurs. As shown in fig. 9, the calculator 500 includes a sampler 601 configured to pass the reception-stage-changed signal Q throughOver a preset delay time to the voltage U of the DC-AC converter 420_DCACThe sampling is performed with a delay time preferably of the order of 50 mus. This delay achieves sampling as close to the maximum voltage as possible. Then, the comparator 602 compares each value U of the voltage according to the following formula_DCAC(n)And its previous value U_DCAC(n-1)And (3) comparison:

if U is present_DCAC(n)>U_DCAC(n-1)+ hysteresis and if U_DCAC(n-1)>U_DCAC(n-2)+ hysteresis, the value of the velocity correction constant K is not equal to 0, otherwise the value of the velocity correction constant K is equal to 0.

Preferably, the value of the speed correction constant K is verified after analyzing several successive samples, in order to confirm the voltage U_DCACEspecially during the operating phases S1-S6_DCACIs measured. Instead of sampling each of the operating phases S1-S6, the voltage U can also be detected by successive sampling with a sufficiently high sampling frequency_DCACAnd thus helps to detect its maximum.

According to a second embodiment, as shown in fig. 4, the current I in the DC-AC converter 420 is detected_DCACAn increase occurs, i.e. an increase in the current between the capacitor of the DC-AC converter 420 and its legs.

Referring to fig. 10, similar to fig. 9, calculator 500 is configured to compare the current I of DC-AC converter 420 with the current I of DC-AC converter_DCACSampling is performed. Then, each current value I is measured_DCAC(n)And compared to the current threshold Iseuil to determine the correction constant K. In this example, the current threshold Iseuil is equal to 0. If the present current I of the DC-AC converter 420_DCAC(n)Negative means that the rotor is decelerating and the speed set point should be increased. Also, if rotor deceleration is detected, the value of the speed correction constant K is set to be not equal to 0, preferably 2% -8%, preferably around 5%.

Thus, the calculator 500 is configured to correct the speed set point in order to optimize the start-up time of the motor 100. In an advantageous manner, fluctuations in the rotor speed are advantageously reduced by the correction.

The following is an embodiment describing a method for starting the synchronous motor 100 according to the present invention with reference to fig. 11.

To rotate the motor 100, the calculator 500 controls the states of the switches T1-T6 according to a drive table shown in fig. 5, which defines successive control phases S1-S6. The rotor acceleration ACC of the motor 100 is preset.

To switch from one control phase S1-S6 to the next control phase S1-S6, the calculator 500 performs a first simple integration of the preset acceleration ACC in step E1, so as to derive the current speed v (n) from the following equation:

V(n)＝V(n-1)+T_S*ACC+W

preferably, during this step, the calculator 500 determines the value of the variable parameter W from the deceleration detection of the electric motor 100 by using one of the previously proposed methods.

The calculator 500 then performs a second simple integration of the resulting velocity v (n) in step E2 to derive therefrom the current electrical angle a (n). The calculator 500 thus generates the electrical angle signal a shown in fig. 6. When the electrical angle signal A reaches the preset threshold angle A_threholdThe phase change signal Q is activated in step E3 in order to switch to the next operating phase S1-S6, which changes the state of the switches T1-T6 (step E4). In addition, the current electrical angle a (n) is reset to zero.

In this manner, the calculator 500 controls the successive changes of the various control phases S1-S6 to achieve the acceleration of the motor 100 to its starting speed.

By means of the method according to the invention, the variation of the control phases S1-S6 is determined from the rotor speed and not from a preset table. This makes it possible to adapt the acceleration to the possible deceleration of the electric motor 10, thus limiting the starting period of the electric motor. As shown in fig. 12, the increase in the estimated velocity VE is rapid. Furthermore, by means of the automatic correction of the speed set-point Vc, the oscillations can be significantly reduced.