DE102018130477A1 - Method for controlling an electric motor - Google Patents

Abstract

A method for controlling an electrically commutated electric motor, the method determining a profile that corresponds to the time course of the rotor position of the electric motor as a result of a setting, determining a maximum target acceleration for a predetermined acceleration profile from the profile, determining a reduced target acceleration, which is smaller than the maximum setpoint acceleration, and includes determining the commutation times of the electric motor for the specified acceleration profile and the reduced setpoint acceleration.

Description

FIELD OF THE INVENTION

The present invention relates to starting processes of electric motors, in particular for the optimization of starting processes in electrically commutated electric motors.

BACKGROUND

Electric motors are used in a wide variety of consumer products and are also taking on an increasing number of tasks as a result of digitalization. However, this increase in electric motors in electronically controlled devices has also led to higher demands being made on individual electric motors with regard to electromagnetic compatibility with other electrical devices and the energy efficiency of the electric motors.

Braking and acceleration processes in particular can lead to an increased electromagnetic interference potential, which can negatively influence the functioning of neighboring electrical devices, and can simultaneously increase the energy consumption of the electric motor. Optimizing and reducing such processes can therefore lead to improved behavior of the electric motor.

In sensorless electric motors, feedback between the movement of the electric motor and the drive, and thus optimal control, is only possible from a certain minimum speed at which the counter-electromotive force, which is generated in the windings of the electric motor by changing magnetic fields, exceeds a detection threshold.

However, such an electric motor is often operated “blindly” below the minimum speed, which may be the reason why starting processes in electric motors are inefficient and slow or lead to a reduced electromagnetic compatibility of the electric motor.

DE 10 2012 102 868 A1 teaches two-stage start-up processes, which consist of a first acceleration phase with pre-calculated commutation times and a second dynamic acceleration phase, in which the electrical commutation is carried out using sensor signals. The acceleration of the motor based on the pre-calculated commutation signals comprises 10 to 15 commutation steps.

DE 10 2013 014 481 A1 also describes two-stage start-up processes in which the pulse width modulation is modified or bridged during the start phase of the electric motor based on the very variable operating DC voltage of the electric motor in order to obtain reliable start-up processes with long switching times of the motor electronics. The acceleration phase of the motor based on the pre-calculated commutation signals comprises 20 commutation steps.

DE 600 25 845 T2 discloses the oversizing of brushless DC motors for operation in the start / stop mode, a varying battery voltage being fully passed on to the windings of the electric motor during acceleration and braking operations and otherwise kept at a predetermined effective voltage via pulse width modulation.

However, both slow start-up processes and the oversizing of the electric motor for start / stop operation do not lead to an efficient and inexpensive solution with good electromagnetic compatibility at the same time.

GENERAL DESCRIPTION OF THE INVENTION

The invention is therefore based on the object of providing improved starting dynamics for electric motors in which the effects of the problems mentioned above are reduced and / or an advantageous compromise can be obtained.

This object is achieved by a method and an electric motor according to the independent claims. The dependent claims relate to preferred embodiments of the invention.

In a first aspect, the invention relates to a method for controlling an electrically commutated electric motor, the method comprising determining a profile that corresponds to the time profile of the rotor position of the electric motor as a result of a setting. The method further comprises determining a maximum target acceleration for a predetermined acceleration profile from the profile, determining a reduced target acceleration which is smaller than the maximum target acceleration, and determining the commutation times of the electric motor for the predetermined acceleration profile and the reduced target acceleration.

It has been found that the maximum target acceleration, which corresponds to the dynamic behavior of the electric motor in relation to a setting, does not achieve optimal behavior in practice. Rather, the choice of commutation steps can be compared to one correspond to the maximum target acceleration reduced acceleration, an improved characteristic can be obtained in sensorless operation of an electric motor. By choosing the reduced target acceleration on the basis of the value of the maximum target acceleration, transient processes of the rotor, which can occur at the beginning of the acceleration profile, can be reduced.

The acceleration profile with the reduced target acceleration can be implemented in the electric motor via a bridge circuit, a phase voltage with a predefined sign being applied to different windings of the electric motor. With cyclical commutation of the states of the switching matrix, a rotation of the rotor can be caused in the electric motor.

The setting can include the phase voltage of the electrically commutated electric motor. The phase voltage can correspond to the voltage amplitude of the voltage which is applied to windings of the electrically commutated electric motor in order to generate a rotation of the electric motor. The voltage can be modulated to obtain a lower effective phase voltage, such as a sinusoidal voltage or pulse width modulated voltage. However, the voltage can always be present with the full amplitude, known as the block voltage, with pulse width modulation being superimposed.

In some embodiments, the effective phase voltage is not reduced in proportion to the reduced target acceleration, at least during the specification of the acceleration profile.

For example, after the reduced target acceleration has been set, the full phase voltage is still used to accelerate the electric motor or the phase voltage is modulated such that the effective phase voltage does not have a value that is reduced in proportion to the reduced target acceleration, but is greater than that reduced in proportion to the reduced target acceleration Is worth. Applying the higher or full phase voltage during acceleration can counteract starting problems of the electric motor under load. In particular, the phase voltage can be set by impressing a phase current. The phase current can be selected, for example, depending on an applied load. For a given input voltage, the phase voltage will consequently be set depending on the impressed current.

In some embodiments, the electric motor is at rest before the setting is made.

The rest position in which the electric motor rests can correspond to a natural rest position of the electric motor, into which the electric motor can be brought by a corresponding setting for a holding operation. The holding operation can include the specification of a holding current which is applied to windings of the electrically commented electric motor. The holding current can be a constant current in the selected windings, the current value of which can be determined via pulse width modulation of the applied phase voltage.

Then the setting can be specified, which specifies an acceleration of the electric motor. For example, starting from a start configuration of the bridge circuit, a block commutation is specified, the phase voltage in the commutated configuration of the bridge circuit being specified up to a first commutation time.

In this way, the profile can provide the dynamic behavior of the electric motor as a result of a setting from the rest position, so that a starting process of the electric motor can be optimized on the basis of the dynamic response of the electric motor to the setting.

The profile can include the measured or simulated time profile of the rotor position or its time derivative as a result of a setting or the model used for the simulation or calculation of the time profile of the rotor position or its time derivative.

In some embodiments, the profile is equal to the time profile of the rotor position or its time derivative.

The time derivative of the rotor position is also called angular velocity in the following. The acceleration of the electric motor can be read, for example, from the time profile of the angular velocity and conclusions can be drawn about the torque of the electric motor.

In some embodiments, the predefined acceleration profile provides for a monotonous, in particular a linear or quadratic, increase in the angular velocity of the electric motor.

The (strictly) monotonous, or linear or quadratic, increase in the wind speed can be specified by the commutation times, the setting specification being able to specify a reduction in the commutation steps.

The dynamic behavior of the electric motor can be read on the basis of the profile, so that an optimal commutation time for accelerating the electric motor can be determined which corresponds to the maximum target acceleration for the specified acceleration profile.

In some embodiments, the method includes determining the optimal commutation time for accelerating the electric motor from the profile, the maximum target acceleration being determined from the optimal commutation time.

In some embodiments, the optimal commutation time corresponds to the commutation time that maximizes the torque of the electric motor and / or minimizes a ripple in the torque of the electric motor during operation.

In some embodiments, the setting specification comprises at least one electrical commutation in the electric motor.

In some embodiments, the setting for determining the profile comprises at least one electrical commutation, in which the phase voltage is provided unmodulated on the windings of the electric motor or, wherein the modulation of the phase voltage generates a predetermined effective phase voltage for the electric motor, such as a lower limit one Battery voltage of a motor vehicle or a power-limiting threshold voltage.

In some embodiments, the determination of the profile as a result of a setting specification comprises a simulation of a controlled system of the electric motor, the controlled system comprising an electrical model of the electric motor and a mechanical model of the electric motor.

The simulation of a controlled system of the electric motor can be implemented on a processing device, such as an ASIC, FPGA or microprocessor, and can be provided as HiL (Hardware In the Loop) or MiL simulation (Model In the Loop). As the output of the simulation model, the dynamic behavior of the controlled system can include the rotor position or its time derivative as a result of the setting as a series of electrical values, the parameters for the simulation of the controlled system of the electric motor representing the properties of the electric motor.

In some embodiments, the profile is measured directly on the electric motor.

In some embodiments, the method further comprises determining the counterelectromotive voltage at the commutation-related detection times based on the profile, comparing the counterelectromotive voltage at the detection times with a detection threshold, and adjusting the target acceleration based on the comparison of the counterelectromotive voltage with the detection threshold.

The counterelectromotive voltage in the electric motor can allow the measurement of the rotor position and thus the sensorless operation of the electric motor. For reliable operation of the electric motor, the counterelectromotive voltage should exceed the detection threshold specific to the electronics of the electric motor.

As soon as the detection threshold value is exceeded, the “blind” operation of the electric motor can be switched over to a feedback operation during a start phase, so that the commutation times can be determined dynamically on the basis of the measured counter-electromotive voltage. In this operation, hereinafter referred to as dynamic control, the electromagnetic compatibility and the energy consumption can be optimized dynamically.

Before switching to dynamic control, an incorrectly selected target acceleration can lead to a predetermined operating speed being exceeded, so that the electric motor must then be braked. Before switching to dynamic control, the detection of the counterelectromotive voltage should also be verified by comparison with expected detection times.

In some embodiments, the maximum target acceleration is reduced in such a way that the electric motor can switch to dynamic control based on the expected detected counter-electromotive voltage before the operating speed is reached.

In some embodiments, the method therefore further comprises obtaining a predetermined operating speed of the electric motor and adapting the target acceleration so that a number of detectable commutation steps in which the counterelectromotive voltage is greater than the detection threshold value is greater than a predetermined minimum value, during one detectable acceleration time window, which corresponds to the number of detectable commutation steps, the electric motor does not reach the operating speed.

In some initial forms, the predetermined minimum value is greater than or equal to 2 or greater than or equal to 3.

Correct detection of the counterelectromotive voltage can be verified on the basis of at least 3 detected commutation steps, so that the motor can be switched to dynamic control before the predetermined operating speed of the electric motor is reached.

In some embodiments, a series of corresponding commutation times is calculated on the basis of the determined maximum target acceleration or the reduced target acceleration, the counterelectromotive voltage being determined for each or at least some of the corresponding commutation times.

On the basis of the determined values of the counterelectromotive voltages, a first detectable commutation time at which the counterelectromotive voltage exceeds the detection threshold value can be determined. The setpoint acceleration can then be adjusted such that the counterelectromotive voltage for the first detectable commutation point in time exceeds the detection threshold value by a predetermined tolerance amount, such as 20% or 10%, or if the counterelectromotive voltage for the first detectable commutation point in time is below that specified by the Tolerance amount is increased detection threshold, the target acceleration is adjusted such that the counterelectromotive voltage for a second detectable commutation time, which follows the first detectable commutation time, exceeds the detection threshold by the predetermined tolerance amount.

The reduced setpoint acceleration can thus be selected such that a high setpoint acceleration can be obtained while simultaneously switching over to dynamic operation as quickly as possible. The tolerance amount can take into account variances of the intended load of the electric motor and / or manufacturing tolerances of the parts of the electric motor.

In some embodiments, the reduced target acceleration is greater than or equal to one third of the maximum target acceleration, in particular greater than or equal to half the maximum target acceleration or approximately equal to half of the maximum target acceleration and less than 80% or 70% of the maximum target acceleration.

A reduced setpoint acceleration, which is greater than or equal to one third of the maximum setpoint acceleration, can favor the electromagnetic compatibility of the electric motor.

In some embodiments, if a reduced target acceleration is less than 40% of the determined maximum target acceleration, the phase voltage is modulated and / or reduced such that the maximum target acceleration for the resulting effective phase voltage does not exceed the reduced target acceleration by a factor that is greater than 2 , 5 is.

In some embodiments, the method further comprises adapting at least the first commutation time determined on the basis of the reduced target acceleration, so that at least the first commutation time includes pre-commutation for the predetermined acceleration profile and the reduced target acceleration.

The determined commutation times for a reduced target acceleration, as determined according to one of the above embodiments, can advantageously be used in electric motors.

In some embodiments, the method further comprises using the commutation times for accelerating an electric motor and / or storing the commutation times in a non-volatile memory of a control unit of the electric motor.

In a further aspect, the invention relates to an electric motor with a non-volatile memory, commutation times which correspond to the determined commutation times according to one of the preceding methods of the first aspect being stored in the non-volatile memory.

The ascertained commutation times can be predetermined during a starting process of the electric motor to accelerate the electric motor from a rest position.

DETAILED DESCRIPTION OF THE EMBODIMENTS

• 1 eine schematische Ansicht eines Elektromotors gemäß einem Beispiel ist;
• 2 einen beispielhaften Regelkreis eines Elektromotors zeigt;
• 3 die Systemkomponenten eines beispielhaften Elektromotors zeigt;
• 4 ein Modell einer Regelstrecke gemäß einem Beispiel zeigt;
• 5 die Komponenten eines mechanischen Teilsystems der Regelstrecke gemäß einem Beispiel veranschaulicht;
• 6 das simulierte dynamische Verhalten eines Elektromotors als Antwort auf eine Stellvorgabe gemäß einem Beispiel zeigt; und
• 7 idealisierte Verläufe der Winkelgeschwindigkeit und der erzeugten gegenelektromotorischen Spannungen für unterschiedliche vorgegebene Beschleunigungen an einem beispielhaften Elektromotor veranschaulicht.

The properties and the various advantages of the method according to the invention can best be seen from the drawings, in which:

• 1 4 is a schematic view of an electric motor according to an example;
• 2nd shows an exemplary control circuit of an electric motor;
• 3rd shows the system components of an exemplary electric motor;
• 4th shows a model of a controlled system according to an example;
• 5 illustrates the components of a mechanical subsystem of the controlled system according to an example;
• 6 shows the simulated dynamic behavior of an electric motor in response to a setting according to an example; and
• 7 Idealized curves of the angular velocity and the counter-electromotive voltages generated for different predetermined accelerations illustrated on an exemplary electric motor.

1 shows an exemplary brushless electric motor 6 with a 3-phase stator 7 with the phase turns W1 , W2 , W3 and a rotor 8th with a pair of poles. The phase turns W1 - W3 each comprise opposing winding sections A , A , B , B , C. , C. on opposite sides of the stator 7 . The pole pair of the rotor 8th includes the magnetic poles N and S .

When a current flows through the turns W1 flows, the winding sections A , Ä together create a magnetic field along their connecting line. In this way, the rotor 8th for example along the magnetic poles N , S with the turn sections A , A , be aligned.

For rotating the rotor 8th should the on the turns W1-W3 applied voltage are cyclically commutated in such a way that a rotating magnetic field generates and thus a torque on the rotor 8th can work. The angular velocity of the rotor can be determined on the basis of the time intervals of the commutations 8th to be controlled.

The rotation of the rotor 8th can over a shaft of the rotor 8th and a transmission connected to the shaft is transmitted to an actuator, so that the electric motor 6 can drive an external load.

During the rotation of the rotor 8th can be the permanent magnet of the rotor 8th in the turns W1-W3 of the stator 7 induce the so-called counter-electromotive voltage (BACK-EMF or BEMF voltage), so that based on the current in the de-energized turns W1-W3 of the stator 7 the position of the rotor 8th can be recognized. This can cause sensorless operation of the electric motor 6 allow, the commutations of the currents in the turns W1-W3 of the electric motor 6 can be carried out based on the induced counter-electromotive voltage.

2nd shows an exemplary structure of a control circuit 10th a block commutation for a sensorless electric motor 6 to the previously described cyclic commutations on the turns W1-W3 applied voltage. The control circuit 10th includes a controller 12 , the bridge circuit from the switches S1-S6 and each to the switches S1-S6 diodes connected in parallel D1-D6 . A phase voltage U _zk is controlled by the control of the electric motor 6 for the control circuit 10th provided and can flow of current through the coil turns W1-W3 of the electric motor 6 effect which with the control circuit 10th are connected.

The regulator 12 can the position of the switch S1-S6 check that the electric motor is running 6 two coil turns each W1-W3 of the electric motor 6 are current-carrying, for those with the switches S1-S6 a predetermined current direction can also be selected. From the different current directions and the different combinations of the turns W1-W3 there are 6 configurations between which the controller 12 can commutate cyclically.

The cyclically commutated, current-carrying turns W1-W3 can generate a rotating magnetic field, which applies a torque to a rotor equipped with permanent magnets 8th causes a controlled rotation of the electric motor 6 can be done.

For example, a six-step block commutation can be implemented, in each of which two of the three phase windings W1 , W2 , W3 are energized and each commutation step, that is to say the time span between two commutation times, corresponds electrically to a rotation of the rotor of 60 °. The semiconductor switches S1 to S6 can also be controlled with a superimposed pulse width modulation (PWM). Typically, a frequency is in the range for the PWM 1 to 100 kHz, especially in the range 10th to 25th kHz selected. This can be done by the electric motor via the pulse duty factor of the PWM 6 caused torque and / or a rotational speed of the rotor 8th , or the actuating speed of an actuator coupled to it, can be set. The commutation steps can then take place electrically at a distance of approximately 60 °. The ideal commutation time is then 30 ° electrically distant in the middle between two BEMF zero crossings. In other configurations, a 12-step commutation can also be carried out, in which two or three phases alternate W1 , W2 , W3 of the electric motor 6 be energized, be provided.

3rd shows the system components for controlling an exemplary electric motor 6 . An engine software 11 can be a controller 14 of the electric motor 6 and the controller 12 to implement. The control 14 can the operating mode and the operating parameters of the electric motor 6 to the controller 12 pretend. The regulator 12 regulates the behavior and receives signals from an actuator 16 , which includes the engine electronics. The engine electronics is with the controlled system 18th of the electric motor 6 coupled so that the actuator 16 about one with the electric motor 6 connected gearbox can control a coupled mechanical load. In addition, the system components can have sensors 20th include which in the controlled system 18th recorded sensor signals to the controller 12 returns so that control of the electric motor 6 optimized and / or the electric motor 6 can be regulated dynamically.

In a sensorless electric motor 6 the controller can perform dynamic control without additional sensors based on the stator windings W1-W3 perform generated counter electromotive voltage in order to dynamically determine optimal commutation times.

During the start phase of the electric motor 6 can change the dynamic position of the rotor 8th due to the low rotational speed of the rotor 8th cannot be determined on the basis of the correspondingly lower counterelectromotive voltage. The commutation of the electric motor can therefore take place in the start phase 6 can be predetermined, the predetermined commutation times can be determined in advance.

4th shows an example of a model of an electric motor 6 for determining a profile of the electric motor 6 , so that a dynamic behavior of the electric motor 6 can be taken into account for the determination of the commutation times. The idealized controlled system 18th includes an electrical subsystem ΣRGS _E and a mechanical subsystem ΣRGS _M which about the internal electrical torque M _i (t) are coupled.

A setting in the form of a by the controller 12 predetermined time-varying voltage curve u (t) on the turns W1-W3 of the electric motor 6 controls the behavior of the electrical subsystem ΣRGS _E . On the mechanical subsystem ΣRGS _M the load moment acts from the outside M _L (t) . The model's initial variables are the angular position φ _m (t) and the angular velocity φ̇ _m (t) of the rotor 8th .

wobei angenommen wird, dass die Windung W1-W3 gleiche Induktivitäten L und Widerstände R besitzen, u₁-u₃ die von dem Regler 12 vorgegebenen Phasenspannungen, i₁-i₃ die Phasenströme in den jeweiligen Windungen W1-W3 und u_p,1-u_p,3 die Polradspannungen sind.The electrical subsystem ΣRGS _E can be described using equation (1): $$\underset{{\overrightarrow{u}}_n}{\underbrace{\left(\begin{array}{l}{u}_1\\ u_{\mathrm{2nd}}\\ u_{\mathrm{3rd}}\end{array}\right)}}=\underset{u_{R,n}}{\underbrace{R\cdot \underset{{\overrightarrow{i}}_n}{\underbrace{\left(\begin{array}{l}{i}_1\\ i_{\mathrm{2nd}}\\ i_{\mathrm{3rd}}\end{array}\right)}}}}+\underset{L_{123}}{\underbrace{\left(\begin{array}{ccc}L& -\frac{L}{\mathrm{2nd}}& -\frac{L}{\mathrm{2nd}}\\ -\frac{L}{\mathrm{2nd}}& L& -\frac{L}{\mathrm{2nd}}\\ -\frac{L}{\mathrm{2nd}}& -\frac{L}{\mathrm{2nd}}& L\end{array}\right)}}\cdot \frac{d}{dt}\underset{{\overrightarrow{i}}_n}{\underbrace{\left(\begin{array}{l}{i}_1\\ i_{\mathrm{2nd}}\\ i_{\mathrm{3rd}}\end{array}\right)}}+\underset{{\overrightarrow{u}}_{p,n}}{\underbrace{\left(\begin{array}{l}{u}_{p\mathrm{,1}}\\ u_{p\mathrm{,\; 2nd}}\\ u_{p\mathrm{,\; 3rd}}\end{array}\right)}}$$

assuming that the swirl W1-W3 same inductors L and resistors R have, u ₁ -u ₃ that of the regulator 12 given phase voltages, i ₁ -i ₃ the phase currents in the respective turns W1-W3 and u _{p, 1} -u _{p, 3} the magnet wheel voltages are.

This results in the phase voltages u ₁ -u ₃ from the sum of the ohmic voltage drops, u _{R, n} , the self-induction voltage due to the inductances L ₁₂₃ of the turns W1-W3 and the magnet wheel voltages u _{p, 1} -u _{p, 3} which of the permanent magnets of the rotor 8th with a rotation in the turns W1-W3 can be generated. Equation (1) can also be written more compactly in vector notation, as shown in equation (2): $${\overrightarrow{u}}_n=R\cdot {\overrightarrow{i}}_n+L_{123}\cdot {\dot{\overrightarrow{i}}}_n+{\overrightarrow{u}}_{p,n}$$

$$u_q=R\cdot i_q+L_d\cdot {\dot{\phi }}_e\cdot i_d+L_q\cdot \frac{d{i}_q}{dt}+{\dot{\phi }}_e\cdot {\widehat{\Psi }}_p$$

wobei R der Widerstand des jeweils aktiven elektrischen Teilsystems, φ̇_e die Winkelgeschwindigkeit des elektrischen 3-Phasen-Systems, L_d und L_q die Längs- und Querkomponenten der Induktivitäten der Windungen W1-W3, i_d , i_q die Längs- und Querkomponenten der Statorströme und Ψ̂_p der von den Permanentmagneten des Rotors 8 erzeugte Polradfluss ist.Using a Clarke transformation, equation (1) can first be converted into a stator-fixed two-dimensional coordinate system and then, for further simplification, into a rotor-fixed coordinate system via a Park transformation, so that equations (3) and (4) can be obtained, which determine the determination the stator voltage u _d along the rotor 8th and the stator voltage u _q across the rotor 8th allow: $$u_d=R\cdot i_d+L_q\cdot {\dot{\phi }}_e\cdot i_q+L_d\cdot \frac{d{i}_d}{dt}$$

$$u_q=R\cdot i_q+L_d\cdot {\dot{\phi }}_e\cdot i_d+L_q\cdot \frac{d{i}_q}{dt}+{\dot{\phi }}_e\cdot {\widehat{\Psi }}_p$$

in which R the resistance of the active electrical subsystem, φ̇ _e the angular velocity of the electrical 3-phase system, L _d and L _q the longitudinal and transverse components of the inductors of the Coils W1-W3 , i _d , i _q the longitudinal and transverse components of the stator currents and Ψ̂ _p those of the permanent magnets of the rotor 8th generated rotor flow.

The thus unshaped voltage equations can then be related to the mechanical subsystem of the controlled system via the torque generated by the stator currents 18th brought to determine the dynamic behavior of the overall system.

5 shows an exemplary model of the mechanical subsystem of the controlled system 18th , which is from the electrical subsystem ΣRGS _E generated internal torque M _i against the frictional torque M _R and the mechanical moment M _M and inertia J _R of the rotor 8th Via the gear ratio i a load with the load torque M _{L, Ab.} and the load mass J _L drives.

The "mechanical" angular position φ _m or angular velocity φ _{̇m of} the rotor 8th depends on the number of pole pairs PP with the "electrical" angular position φ _e or angular velocity φ _{̇e of} the electrical 3-phase system in a known manner and causes a movement of the position via the gear ratio i φ _L of the actuator with the angular velocity φ _̇L .

wobei zur Vereinfachung das Massenträgheitsmoment des Antriebssystems als Trägheitsmoment $J_M=J_R+\frac{J_L}{i^2}$

zusammengefasst wurde und wobei $M_L=\frac{M_{L,Ab.}}{i}$

das effektiv von dem Getriebe mit der Getriebeübersetzung i an den Elektromotor 6 übertragene Lastmoment der mechanischen Last ist.For the coupled mechanical and electrical subsystem, the time derivative of the electrical angular velocity and thus the acceleration φ̈ _e can be determined using equation (5): $${\ddot{\phi }}_e=PP\cdot \frac{\left(M_i-M_R-M_L\right)}{J_M}$$

where, for simplification, the moment of inertia of the drive system as the moment of inertia $J_M=J_R+\frac{J_L}{i^{\mathrm{2nd}}}$

was summarized and where $M_L=\frac{M_{L,Ab.}}{i}$

that effectively from the gearbox with the gear ratio i to the electric motor 6 transmitted load moment of the mechanical load.

wobei x₁ die Längs-Stromkomponente i_d , x₂ die Quer-Stromkomponente i_q , x₃ die elektrische Winkelgeschwindigkeit φ̇_e, x₄ die elektrischen Winkelposition φ_e bezeichnet und d_m die Reibungskonstante ist.The system of differential equations of the overall system according to equation (6) can then be obtained from equations (3), (4) and (5): $$\begin{array}{c}\left(\begin{array}{l}{\dot{x}}_1\\ {\dot{x}}_{\mathrm{2nd}}\\ {\dot{x}}_{\mathrm{3rd}}\\ {\dot{x}}_{\mathrm{4th}}\end{array}\right)=\left(\begin{array}{cccc}-\frac{R}{L_d}& \frac{L_q}{L_d}\cdot x_{\mathrm{<figure-callout\; id="0"\; label="3rd"\; filenames="DE102018130477A1\_0019.png,DE102018130477A1\_0020.png"\; state="\{\{state\}\}">3rd</figure-callout>}}& 0& 0\\ -\frac{L_q}{L_d}\cdot x_{\mathrm{3rd}}& -\frac{R}{L_d}& -\frac{{\dot{\Psi }}_{\mathrm{<figure-callout\; id="0"\; label="p\; L\; q"\; filenames="DE102018130477A1\_0019.png,DE102018130477A1\_0020.png"\; state="\{\{state\}\}">p\; </figure-callout>}}}{L_q}& 0\\ \frac{\mathrm{3rd}}{\mathrm{2nd}}\frac{P{P}^{\mathrm{2nd}}\cdot \left(L_q-L_d\right)}{J_M}\cdot x_{\mathrm{2nd}}& \frac{\mathrm{3rd}}{\mathrm{2nd}}\frac{P{P}^{\mathrm{2nd}}{\dot{\Psi }}_p}{J_M}& -\frac{d_{\mathrm{<figure-callout\; id="0"\; label="m\; J\; M"\; filenames="DE102018130477A1\_0019.png,DE102018130477A1\_0020.png"\; state="\{\{state\}\}">m\; </figure-callout>}}}{J_M}& 0\\ 0& 0& 1& 0\end{array}\right)\cdot \left(\begin{array}{l}{x}_1\\ x_{\mathrm{2nd}}\\ x_{\mathrm{3rd}}\\ x_{\mathrm{4th}}\end{array}\right)+\left(\begin{array}{cc}\frac{1}{{\mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="DE102018130477A1\_0019.png,DE102018130477A1\_0020.png"\; state="\{\{state\}\}">L\; </figure-callout>}}_d}& 0\\ 0& \frac{1}{{\mathrm{<figure-callout\; id="0"\; label="L\; d"\; filenames="DE102018130477A1\_0019.png,DE102018130477A1\_0020.png"\; state="\{\{state\}\}">L\; </figure-callout>}}_d}\\ 0& 0\\ 0& 0\end{array}\right)\cdot \left(\begin{array}{l}{U}_d\\ U_q\end{array}\right)\\ +\left(\begin{array}{c}0\\ 0\\ -\frac{PP}{J_M}\\ 0\end{array}\right)\cdot M_L\end{array}$$

where x _{1 is} the longitudinal current component i _d , x ₂ the cross current component i _q , x ₃ the electrical angular velocity φ̇ _e , x ₄ the electrical angular position φ _e designated and d _m is the friction constant.

After appropriate characterization of the electric motor 6 can therefore, in principle, the dynamic behavior of the actuator as a function of a setting by appropriate magnet wheel voltages u _d , u _q be preserved. Thus, a profile of the electric motor 6 for a specified setting, taking into account the electrical and magnetic properties of the rotor / stator combination, the friction, the external mechanical load, and / or the moments of inertia of the rotor 8th and the mechanical subsystem can be determined. The setting should be the acceleration process during the start phase of the electric motor 6 replicate.

wobei f_sum die Summe aller vorherigen Kommutierungszeitpunkte t_k(t) ist, sodass die Abstände zwischen Kommutierungen während der Beschleunigung des Elektromotors 6 kontinuierlich geringer werden.To accelerate an electric motor 6 from the rest position during a starting phase can on the turns W1-W3 of the electric motor 6 the phase voltage U _zk applied and after a certain time the current-carrying turns W1-W3 be commutated. Here, an acceleration profile of the electric motor 6 be specified as a linear acceleration profile so that the wind speed of the electric motor 6 increases linearly. For a given acceleration α, the time of the next commutation point in time t _k can be determined using equation (7): $$t_k\left(t\right)=-\frac{t_{sum}}{\mathrm{2nd}}\cdot \sqrt{{\left(\frac{t_{sum}}{\mathrm{2nd}}\right)}^{\mathrm{2nd}}+\frac{\mathrm{10th}}{a\cdot PP}}$$

in which f _sum the sum of all previous commutation times t _k (t) is so that the distances between commutations during acceleration of the electric motor 6 continuously decrease.

If the acceleration process is preceded by a stop operation, the rotor can 8th based on the holding current, and it must until the new commutation position is reached, turn only 30 ° electrically instead of 60 ° electrically, as with a conventional six-step commutation or 60 ° block commutation in stationary operation. The rotor 8th can take the new commutation position after only 50% of the time provided for this. The acceleration profile can therefore provide for a corresponding shortening of the first commutation step in the event of a previous stop operation.

$$\frac{dz}{dt}=v-{\sigma }_0\frac{v}{g\left(v\right)}z$$

A dynamic friction model, for example the LuGren model, can also be considered to determine the commutation times. The LuGren model is based on the approach that the friction as a resistance force is deflected by many small bristles which, as soon as external forces act on a body, cause a counterforce. The LuGren model considers the sum of all bristles in the form of a single bristle that describes the aggregated state of all bristles. A first order nonlinear differential equation is created: $$F={\sigma }_{\mathrm{e.g.}}+{\sigma }_1\left(v\right)\frac{d\mathrm{e.g.}}{dt}+{\sigma }_{\mathrm{2nd}}\left(v\right)$$

$$\frac{d\mathrm{e.g.}}{dt}=v-{\sigma }_0\frac{v}{G\left(v\right)}\mathrm{e.g.}$$

Equation (8) corresponds to a damped oscillation equation with a spring constant or spring stiffness σ0 and a damping function σ1 (v). The damping function σ1 (v) can be selected as follows, for example: $${\sigma }_1\left(v\right)={\sigma }_1{e}^{-{\left(\frac{v}{v_d}\right)}^{\mathrm{2nd}}}$$

Here is σ1 a damping constant and v _D a parameter dependent on the speed v. After Canudas de Wit, H. Olsen KJ Astrom, and P. Lischinsky. "A new model for control of system with friction", IEEE TRANSACTION ON AUTOMATIC CONTROL, 40, No. 3, 1995, the function g (v) can be selected as follows: $$G\left(v\right)=\frac{1}{{\sigma }_0}\left(F_{\mathrm{C.}}+\left(F_H-F_{\mathrm{C.}}\right)e^{-{\left(\frac{v}{v_S}\right)}^{\mathrm{2nd}}}\right)$$

g (v) describes the transition from static to sliding friction, the transition using the parameter v _p can be modeled. Alternatively, the transition from static to sliding friction can also be modeled differently.

wobei δ_v ein Parameter ist, der üblicherweise im Bereich 1 ≤ δ_v ≤ 2 gewählt wird. Für positive Geschwindigkeiten und δ_v=1 ergibt sich folglich die übliche Formel der Stoke'schen (viskosen) Reibung: $${\sigma }_2\left(v\right)=\sigma v\text{.}$$

Means σ ₂ (v) a portion of viscous friction can also be taken into account with $${\sigma }_{\mathrm{2nd}}\left(v\right)=\sigma {\left|v\right|}^{{\delta }_v}siGn\left(v\right),$$

where δ _{v is} a parameter that is usually selected in the range 1 δ _v 2 2. For positive velocities and δ _v = 1 the usual formula of Stoke's (viscous) friction results: $${\sigma }_{\mathrm{2nd}}\left(v\right)=\sigma v\text{.}$$

6 shows a corresponding course of the angular velocity of the rotor 8th of a simulated exemplary electric motor 6 as a function of time t with a given acceleration a, with a setting in the form of a cyclic commutation of the magnet wheel voltages u _{p, 1} , u _{p, 2nd} , u _{p, 3} with the voltage amplitude U _zk in the three phases of an electric motor 6 without external load moment M _L is specified. The simulated course can be simulated by the dynamic behavior of the exemplary electric motor 6 can be obtained according to equation (6).

At time t = 0.0037 s, a first commutation is carried out so that the electric motor 6 is accelerated from its rest position. A second commutation is carried out at t = 0.012 s.

The mechanical angular velocity φ̇ _{m of} the electric motor 6 is accelerated to t = 0.0065 s to an angular velocity of φ̇ _m = 1000 rpm, but then decelerated to t = 0.0085 s and then to t = 0.0011 s in the opposite direction of rotation to an angular velocity of φ̇ _m = - 800 rpm accelerates.

Thus the in 6 shown electric motor 6 too fast with regard to the external setting, so that the rotor swings over and back 8th regarding the stator 7 takes place. This sequence of decelerations and accelerations can then continue until the time at which the commutation steps are close enough in time to determine the dynamic behavior of the electric motor 6 approximate. However, this settling process before the effective acceleration can have an electromagnetic compatibility of the electric motor 6 reduce and excessive stress and high energy consumption of the electric motor 6 to lead.

A profile of the electric motor 6 like it in 6 is illustrated, however, may allow a maximum target acceleration α to determine which is the dynamic behavior of the electric motor 6 corresponds. For this purpose, the optimal commutation time or a corresponding acceleration can be read from the course of the angular velocity and then the target acceleration of the electric motor 6 be adjusted. The reduced target acceleration of the electric motor 6 can the dynamic behavior of the electric motor 6 map and can the occurrence of a sequence of acceleration and deceleration processes of the rotor 8th minimize.

For example, the maximum target acceleration can be calculated from the optimal commutation time of t = 0.0045 s 6 corresponding profile of the electric motor 6 be determined.

At the same time, however, it has been shown that the specification of the maximum target acceleration in connection with low operating speeds, fluctuating load torques of the electric motor 6 or manufacturing tolerances do not always lead to optimal results. In particular, in order to optimize the starting phase, the transition between “blind” operation and dynamic control can be taken into account on the basis of the counterelectromotive voltage.

7 shows idealized speed profiles (solid lines) and corresponding ones in the turns W1-W3 induced counterelectromotive voltages (BEMF, dashed lines) each as a function of the switching points, which are the times for commutation of the switches S1-S6 of the control circuit 10th represent, for an exemplary simulated behavior of an electric motor 6 with an additional external load moment M _L . The dashed horizontal line marked with diamonds marks the detection threshold of the electric motor 6 to the counter-electromotive voltage BEMF in the control electronics of the electric motor examined as an example 6 to be able to reliably detect.

The speed curve and the corresponding counter-electromotive voltage BEMF for the maximum target acceleration of the exemplary electric motor 6 of α = 2317 1 / s ² is marked by circles and shows a rapid increase in speed. The exemplary illustrated electric motor 6 is designed for an operating speed of 1500 rpm, which is exceeded at the maximum target acceleration after the second switching point.

At this point in time, however, only two signals of the counter electromotive voltage could be generated by the control electronics of the electric motor 6 can be detected. However, control electronics that are less prone to errors often require the detection of at least three signals from the counterelectromotive voltage in order to switch over to dynamic regulation of the commutation times on the basis of the counterelectromotive voltage. At the time of detection of the third signal induced by the counter electromotive voltage, the electric motor is 6 however, already above the operating speed and must then be braked during dynamic control.

An optimized start phase is also in 7 shown. The corresponding speed curve and the counter-electromotive voltage BEMF for the reduced target acceleration of the exemplary electric motor 6 of α = 1250 1 / s ² is characterized by triangles, the target acceleration being reduced to approximately half the maximum target acceleration. In particular, a reduced target acceleration between 800 m / s ² and 1700 m / s ² , or 900 m / s ² and 1600 m / s ² , preferably between 1000 m / s ² and 1500 m / s ^{2 can be} selected if the maximum Target acceleration is approximately a = 2300 m / s ² .

In other words, the ratio between reduced target acceleration and maximum target acceleration should be chosen between values of 35% and 75% or between values of 40% and 70% or between values of 45% and 65% in order to improve the behavior of the electric motor 6 to obtain.

In the present example, the reduced target acceleration was reduced to approximately half with respect to the maximum target acceleration, so that a number of sensor signals of the counterelectromotive voltage, which is greater than two, can be detected before the operating speed is exceeded. However, since the reduced target acceleration is greater than a third of the maximum target acceleration, it occurs during the acceleration of the electric motor 6 not a reversal of the angular velocity, so that the electromagnetic compatibility and also the energy consumption of the electric motor 6 can be improved.

In this case, at least for the first commutation time, a pre-commutation of greater than 15 °, such as 30 ° (with respect to the electrical angular position), can be selected, ie the first commutation is carried out earlier than the intended time, the position of the electric motor 6 at the time of Pre-commutation is shifted by 30 ° relative to the intended position, for example to prevent a reversal of the angular velocity. The pre-commutation can be continued for the following commutation times, it being possible to reduce an angle corresponding to the pre-commutation, for example monotonously with the number of commutations.

The reduced target acceleration is preferably adjusted such that the counterelectromotive voltage at the first commutation-related detection time exceeds the detection threshold value by a predetermined tolerance amount, such as 20% or 10%. In this way, reliable detection of the first intended counter electromotive voltage provided can be ensured. Such an adjustment can affect the operation of the electric motor 6 improve at low operating speeds.

In addition, a reduction and / or modulation of the phase voltage U _ZK be carried out for the starting phase, if the target acceleration is reduced according to a profile of the electric motor, as before 6 a swinging back of the rotor for the reduced target acceleration 8th is determined.

Thus, by reducing the maximum target acceleration to a reduced target acceleration, an advantageous compromise for the operation of the electric motor 6 be preserved.

Reference symbol list

6

Electric motor

7

stator

8th

rotor

10th

Control circuit

11

Engine software

12th

Regulator

14

control

16

Actuator

18th

Controlled system

20th

Sensors

W ₁ -W ₃

Pairs of turns

N, S

Magnetic poles

S ₁ -S ₆

counter

D ₁ -D ₆

Diodes

U _zk

Phase voltage

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102012102868 A1 [0006]
• DE 102013014481 A1 [0007]
• DE 60025845 T2 [0008]

Claims (17)

A method of controlling an electrically commutated electric motor, the method comprising: Determining a profile which corresponds to the time course of the rotor position of the electric motor as a result of a setting, Determining a maximum target acceleration for a predetermined acceleration profile from the profile, - Setting a reduced target acceleration, which is less than the maximum target acceleration, and - Determining the commutation times of the electric motor for the given acceleration profile and the reduced target acceleration. Procedure according to Claim 1 , the profile being equal to the time course of the rotor position or its time derivative. Method according to one of the preceding claims, wherein the electric motor rests before the setting. Method according to one of the preceding claims, wherein the setting specification comprises at least one electrical commutation in the electric motor. Method according to one of the preceding claims, wherein the determination of the profile as a result of a setting specification comprises a simulation of a controlled system of the electric motor, and wherein the controlled system comprises an electrical model of the electric motor and a mechanical model of the electric motor. Method according to one of the preceding claims, wherein the predetermined acceleration profile provides a monotonous, in particular linear or quadratic, increase in the angular velocity of the electric motor. The method of any one of the preceding claims, further comprising: - Determining an optimal commutation time for accelerating the electric motor from the profile, the maximum target acceleration being determined from the optimal commutation time. Procedure according to Claim 7 , wherein the optimal commutation time corresponds to the commutation time, which maximizes the torque of the electric motor and / or minimizes a ripple in the torque of the electric motor during operation. The method of any one of the preceding claims, further comprising: - Determining the counterelectromotive voltage at the commutation-related detection times using the profile, and Comparing the counterelectromotive voltage at the detection times with a detection threshold value, - Adjusting the target acceleration based on the comparison of the counterelectromotive voltage with the detection threshold. Procedure according to Claim 9 , which further comprises: - obtaining a predetermined operating speed of the electric motor, - adjusting the maximum target acceleration, - so that a number of detectable commutation steps in which the counter-electromotive voltage is greater than the detection threshold value is greater than a predetermined minimum value, - during one detectable acceleration time window, which corresponds to the number of detectable commutation steps, the electric motor does not reach the operating speed. Procedure according to Claim 10 , the predetermined minimum value being greater than or equal to 3. Method according to one of the preceding claims, wherein the reduced target acceleration is greater than or equal to one third of the maximum target acceleration, in particular greater than or equal to half the maximum target acceleration or approximately equal to half of the maximum target acceleration. A method according to any one of the preceding claims, the method further comprising: - Using the commutation times to accelerate an electric motor and / or - Storage of the commutation times in a non-volatile memory of a control unit of the electric motor. Method according to one of the preceding claims, wherein the method is carried out on a sensorless electric motor. The method of any one of the preceding claims, further comprising: - Adjusting at least the first commutation time of the determined commutation times, so that at least the first commutation time includes pre-commutation for the predetermined acceleration profile and the reduced target acceleration. Electric motor with a non-volatile memory, being in the non-volatile memory Commutation times are stored which correspond to the determined commutation times according to one of the preceding methods Claims 1 to 14 correspond. Electric motor with a processor, which is set up, the method according to the Claims 1 to 15 to obtain commutation times for the acceleration of the electric motor.

Images