阅读说明：本技术 用于在电机的方波操控和pwm操控之间切换的方法 (Method for switching between square-wave and PWM actuation of an electric machine ) 是由 D·塞潘斯基 于 2018-08-09 设计创作，主要内容包括：本发明涉及一种用于运行电机的方法,所述电机能够借助PWM操控(A<Sub>1</Sub>)和方波操控(A<Sub>3</Sub>)来运行,其中将过渡操控(A<Sub>2</Sub>)用于在PWM操控(A<Sub>1</Sub>)和方波操控(A<Sub>3</Sub>)之间的过渡,其中在电机转矩调节的范围中将相电压的d值设定为调节变量,并且持续地改变相电压的q值。(The invention relates to a method for operating an electric machine that can be controlled (A) by means of PWM 1 ) And square wave manipulation (A) 3 ) To operate, wherein the transition is manipulated (A) 2 ) For controlling in PWM (A) 1 ) And square wave manipulation (A) 3 ) In which the d value of the phase voltage is set as a manipulated variable in the range of the motor torque control and the q value of the phase voltage is continuously changed.)

1. Method for operating an electric machine (100) which can be controlled (A) by means of PWM₁) And square wave manipulation (A)₃) To be operated in a manner such that,

wherein the transition is manipulated (A)₂) For controlling (A) in said PWM₁) And said square wave manipulation (A)₃) In a range of regulation of the torque (M) of the electric machine (100), a phase voltage (U) is applied to the phase voltage_d) Is set as a regulating variable and the phase voltage (U) is continuously varied_q) Q value of (2).

2. The method of claim 1, wherein in slave to the PWM manipulation (a)₁) Transition to the square wave manipulation (A)₃) While increasing the phase voltage (U)_q) And/or said q value of

Wherein in the steering from the square wave (A)₃) Transition to the PWM manipulation (A)₁) While decreasing the phase voltage (U)_q) Said q value of (1).

3. Method according to claim 1 or 2, wherein the phase voltage (Uc) is varied with a preset gradient (AU)_q) Said q value of (1).

4. The method of any of the preceding claims, wherein (A) is performed from a previous manipulation₁，A₃) Starting from the last used value, the phase voltage (U) is changed_q) Said q value of (1).

5. Method according to any of the preceding claims, wherein the pole rotor voltage (U) of the electrical machine when a transition is started_P) Is greater than or equal to the intermediate circuit voltage (U) of the motor_dc) Multiplying by a factor 2/Pi, preferably in each case adding a presettable offset (O) which is positive or negative depending on the direction of transition, directly and without the transition control carrying out the secondary PWM control (A)₁) Transition to the square wave manipulation (A)₃) And/or wherein the pole-rotor voltage (U) of the machine is at the beginning of the transition_P) Less than or equal to the intermediate circuit voltage (U) of the motor_dc) Multiplying by a factor 2/Pi, preferably adding a presettable offset (O) which is positive or negative depending on the transition direction, respectively, directly and without the transition control performing the control (A) from the square wave₃) Transition to the PWM manipulation (A)₁)。

6. Method according to any of the preceding claims, wherein the rotational speed (n) of the electric machine is greater than a predefinable rotational speed threshold (n)_Gr) Preferably, the slave PWM control (A) is carried out with a presettable positive offset (O)₁) Switching to the transition maneuver (A)₃)。

7. The method of any one of the preceding claims, wherein the method is carried out whileSaid phase voltage (U)_dq) An absolute value in a d-q coordinate system equal to or greater than the intermediate loop voltage (U)_dc) Performing a transition maneuver (A) from the transition when multiplied by a factor 2/Pi₂) Switching to said square wave manipulation (A)₃)。

8. Method according to any of the preceding claims, wherein the rotational speed (n) of the electric machine is less than a pre-settable rotational speed threshold (n)_Gr) Preferably, when a preset positive deviation (O) is reduced, a control (A) from the square wave is performed₃) Switching to the transition maneuver (A)₂)。

9. Method according to any of the preceding claims, wherein the current-time phase (I) is_d) Is equal to or less than a reference value (I) of the d value of the phase current in the PWM manipulation_d,ref) Preferably, when a preset positive deviation (O) is reduced, a transition operation (A) is executed₂) Switching to the PWM manipulation (A)₁)。

10. A computing unit (140) for performing the method according to any of the preceding claims.

11. A computer program which, when run on the computing unit (140), causes the computing unit (140) to perform the method according to any one of claims 1 to 9.

12. A machine-readable storage medium having stored thereon a computer program according to claim 11.

Technical Field

The invention relates to a method for operating an electric machine, and to a computing unit and a computer program for carrying out the method.

Background

Electrical machines, in particular generators, can be used to convert mechanical energy in a motor vehicle into electrical energy. For this purpose, claw-pole generators are usually used, which are usually equipped with an electrical excitation mechanism. Since claw pole generators produce polyphase, usually three-phase, currents, a commutation is necessary for conventional motor vehicle dc voltage mains. Rectifiers based on semiconductor diodes or semiconductor switches can be used for this purpose.

The generator may also be used to start the internal combustion engine. Such a generator is also referred to as a starter generator. Such starter generators are usually operated with only very low rotational speeds, since the torque that can be generated decreases rapidly with respect to the rotational speed. However, larger electric machines are also conceivable, which can also be used for driving the vehicle in a hybrid vehicle, or at least for assisting the internal combustion engine.

For controlling such starter generators, so-called PWM (pulse width modulation) operation, in which the phase currents are regulated, or so-called square wave (Block) control, in which the advance commutation angle can be varied, can be used. The two types of actuation can be switched between according to a rotational speed threshold.

Disclosure of Invention

According to the invention, a method for operating an electric machine, as well as a computing unit and a computer program for carrying out the method are proposed with the features of the independent claims. Advantageous embodiments are the subject of the dependent claims and the following description.

The method according to the invention is used for operating an electric machine, for example a claw pole machine, which can be operated by means of PWM control and by means of square wave control. So-called field-oriented regulation is preferably used in PWM control or PWM operation. The phase currents are measured and converted into a d-q coordinate system or into a so-called space vector representation. The parameters of the electric machine can be calculated via a fitting polynomial which describes a saturation dependence and from which a reference current can then be calculated, which can then be converted into a voltage reference value (in the d-q coordinate system) via field-oriented regulation. Subsequently, it can be converted into a pulse pattern (e.g. a so-called center-aligned pulse pattern) in the PWM controller and regulated at the phases of the electric machine via inverters or converters.

In the case of square wave (Block) actuation or square wave (Block) operation, the so-called advance commutation angle can be set or preset. The leading commutation angle provides when a semiconductor switch connected to a phase is switched into the conducting state with regard to the passage of the zero point of the pole rotor voltage induced in a phase. Instead of regulating the current, the current torque can be calculated from the machine parameters and the current. The current torque can then be compared with the setpoint torque and the lead commutation angle can be adjusted based on the deviation (e.g. using a PI controller). The resulting leading commutation angle can then be converted directly into a square-wave pattern by always feeding the largest phase voltage. The lead commutation angle can be understood in the d-q coordinate system as the angle between the q value and the d value of the phase voltage.

In general, switching from PWM control to square-wave control takes place when the rotational speed of the electric machine exceeds a predetermined rotational speed threshold or when the pole rotor voltage exceeds the intermediate circuit voltage. The switching from square-wave operation to PWM operation takes place in reverse and, if necessary, hysteresis is present in order to prevent a continuous switching back and forth of the operation type. The rotational speed threshold (the regulation used usually does not operate until it is stable) can be correlated to the processor used or to its speed. The pole rotor voltage is generated by the rotation of the rotor or rotor and is based on induction in the winding branches or phases.

In the case of a voltage step in the reference preset of the phase voltages, which does not occur during the switching due to the ratio of the pole rotor voltage to the intermediate circuit voltage when the intermediate circuit voltage is always completely applied in the square-wave control, it can now occur during the switching due to the speed threshold value, for example, in order to ensure the control stability, that the pole rotor voltage is lower than the intermediate circuit voltage. This voltage step causes a step in the phase voltage and thus in the torque of the motor.

In order to transition between PWM control and square-wave control, it is now possible according to the proposed method to use transition control, in which the d value of the phase voltage (phase voltage is therefore used here in the d-q coordinate system) is set as a control variable in the framework of the motor torque control, and the q value of the phase voltage is continuously, in particular (quasi-) continuously, changed. The q value of the phase voltage is preferably increased when transitioning from the PWM control to the square wave control, and the q value of the phase voltage is preferably decreased when transitioning from the square wave control to the PWM control. In this case, the q value can be varied with a predetermined gradient. Here, the value last used in the previous operation (i.e., PWM operation or square wave operation) is considered as a starting point of the change of the q value of the phase voltage, as appropriate.

Instead of initiating a voltage step, the q value of the phase voltage is changed only slowly by this transition control, wherein the d value remains substantially constant. Since the d value has a significant influence on the q value of the phase current, said q value remains substantially unchanged. The q value of the phase currents in turn has a significant influence on the torque of the electric machine, which therefore likewise remains substantially constant.

Advantageously, the transition from the PWM control to the square-wave control is carried out directly and without transition control if the pole-rotor voltage of the electric machine at the beginning of the transition is greater than or equal to the intermediate circuit voltage of the electric machine multiplied by a factor 2/Pi. The transition from square-wave operation to PWM operation is preferably carried out directly and without transition operation when the pole-rotor voltage of the electric machine at the beginning of the transition is less than or equal to the intermediate circuit voltage of the electric machine multiplied by a factor 2/Pi. In this case, a deviation that can be preset, positive or negative depending on the transition direction, is preferably taken into account. In this case, no voltage step occurs in the direct transition, so that no transition control is required. However, certain voltage steps of, for example, a maximum of 0.5V can also be tolerated. The natural oscillation, which is converted into the maximally adjustable sinusoidal form in the square-wave operation, is realized with a factor 2/Pi.

Advantageously, the switch from the PWM control to the transition control is carried out when the rotational speed of the electric machine is greater than a predefinable rotational speed threshold value, preferably plus a predefinable positive deviation. In this case, therefore, it is not based on the pole rotor voltage being greater than or equal to the intermediate circuit voltage, so that voltage steps can be prevented with transient actuation.

Advantageously, the switchover from transition control to square-wave control is carried out when the absolute value of the phase voltage, in particular the internal reference phase voltage in the d-q coordinate system, is equal to or greater than the intermediate circuit voltage multiplied by a factor 2/Pi. The voltage is then increased by the transition control, so that no voltage step and therefore no torque jump occurs.

Preferably, the switching from square-wave operation to transition operation is carried out when the rotational speed of the electric motor is less than a predefinable rotational speed threshold, preferably minus a predefinable positive deviation. In this case, it is therefore also not possible to base the pole rotor voltage also on or above the intermediate circuit voltage, so that voltage steps can be prevented by means of transient actuation.

Advantageously, the switching from the transition control to the PWM control is performed when the d value of the phase current is equal to or smaller than the reference value of the d value of the phase current in the PWM control, preferably minus a preset positive deviation. Since the value of the voltage is not known in advance here, this can be achieved by the conditions mentioned for the current, and no voltage or torque step occurs.

The deviation can be used in the proposed case in terms of hysteresis to prevent a continuous switching between the types of actuation at the limits.

The computing unit according to the invention (for example, a control unit of a motor vehicle) is provided in particular in terms of programming for carrying out the method according to the invention.

It is also advantageous to execute the method in the form of a computer program, since this results in a particularly low cost, especially when the execution controller is also used for other tasks and is therefore necessarily present. Suitable data carriers for supplying the computer program are in particular magnetic, optical and electrical memories, such as hard disks, flash memories, EEPROMs, DVDs etc. The program may also be downloaded via a computer network (Internet, Intranet, etc.).

Further advantages and embodiments of the invention are given in the description and the figures.

Drawings

The invention is schematically illustrated in the drawings and will be described hereinafter with reference to the drawings according to embodiments.

Fig. 1 schematically shows an electrical machine in which the method according to the invention can be carried out.

Fig. 2 schematically shows a variant of the control of the electric machine.

Fig. 3 schematically shows a current curve in an electric machine.

Fig. 4 shows the phase voltages of the phases at the transition between a PWM control and a square wave (Block) control.

Fig. 5a and 5b schematically show the range of different types of manipulation.

Fig. 6 shows different variants of an electrical machine not using the method according to the invention.

Fig. 7, 8 and 9 show three different actuation types, as can be used in a preferred embodiment in the method according to the invention.

Fig. 10 shows a flow chart of a method according to the invention in a further preferred embodiment.

Fig. 11 shows different variants of the electric machine when using the method according to the invention in another preferred embodiment.

Detailed Description

Fig. 1 shows a schematic circuit diagram of an electric machine in which the method according to the invention can be carried out. The electric machine 100 is designed here, for example, as an externally excited five-phase electric machine. Of course, a different number of phases, for example three, can also be used. Furthermore, the electric machine 100 can be designed, for example, as a claw pole electric machine.

In this case, the electrical machine 100 has five stator windings 120 and one field or rotor winding 110. The field current I in the field winding 110 can be set by a computing unit designed as a control unit 140_ex. Furthermore, a circuit arrangement 130 is provided, which has switches 131, here for example MOSFETs, only one of which is identified by a reference numeral, by means of which a voltage U can be applied to or tapped off from the stator windings 120, depending on whether the electric machine is used for motor operation or generator operation.

The switching device 130 and the control unit 140 may also be part of a common control unit for the motor or part of an inverter for the motor, respectively.

Fig. 2 schematically shows a solution for operating an electric machine such as that shown in fig. 1. The upper graph shows the voltage U as a function of time T and the lower graph shows the duty cycle T as a function of time T.

Here, a steering mode according to the standard method of so-called triangle-sine modulation is involved. The desired target voltage, i.e. the sine curve in the upper diagram, is superimposed by a triangular signal (also shown in the upper diagram) which has a significantly higher frequency than the triangular signal of the fundamental wave (typically greater than 10 kHz). Each crossing marks a switching of the PWM signal.

The steering model in the lower graph may be generated using a PWM controller. For claw pole motors, this PWM control is typically used without exceeding the voltage limit. From or above the voltage limit, the motor is then operated using a square wave (Block) model running with a square wave (Block) or using a so-called square wave (Block) control.

This square-wave (Block) manipulation is characterized in that the phase voltages have the largest possible amplitude and this amplitude cannot be changed (this can theoretically be achieved by manipulation with a square-wave (Block) width of less than 180 °, but is not generally used).

In addition to the excitation current, the phase position of the voltage vector, the so-called leading commutation angle, is also used as a manipulated variable for the desired setpoint torque of the electric machine.

Since the amplitude cannot be modified (or at least not) in the case of square wave (Block) operation, this control mode is used only when the voltage limit is exceeded (i.e. when the pole rotor voltage is greater than the applied intermediate circuit voltage), which is typically used in the present case at rotational speeds greater than 3000 a/min.

Fig. 3 shows the current profile, in this case the phase currents, which are generated when the electric machine is rotating and sinusoidal phase voltages are specified. For this purpose, the current I is plotted as a function of time t.

In the case of a symmetrical distribution, the individual phase currentsI_A、I_B、I_C、I_DAnd I_EA space vector may be synthesized. Known Clarke (Clarke) and Park transforms are used for this purpose. First, the current I is obtained by Clarke transformation according to the following formula_αAnd I_β：

And

then, the current I is obtained by Park conversion according to the following formula_dAnd I_q：

I_d＝I_αcosθ+I_βsin theta and I_q＝-I_αsinθ+I_βcosθ

Where theta denotes the angle of the rotor or rotor of the motor. In the same way, the voltage can also be transformed into the d-q coordinate system.

Fig. 4 shows the phase voltages of the phases at the transition between PWM control and square wave (Block) control or square wave (Block) operation. For this purpose, the voltage U and the current I are plotted as a function of time t. As can be seen from the current curve, the phase current is no longer regulated here.

In fig. 5a and 5b, the range of different actuation types is shown, according to which the control strategy in the current method is to be set forth. For this purpose, the torques M are plotted in each case with respect to the rotational speed n of the electric machine.

Here, the range I describes a range in which the pole rotor voltage is also lower than the intermediate circuit voltage. Range III specifies the range in which rotational speed n is greater than rotational speed threshold n_Gr。

For higher torques, in the externally excited electric machine, higher excitation currents are set in each case such that the pole rotor voltage is below the limit speed n_GrIn the case of (2) exceeds the intermediate loop voltage. For constant excitation current, pole rotor voltage becomesThe ratio is increased.

The limit speed n here_GrThe rotational speed threshold value up to the steady operation of the phase current regulation is indicated. The rotational speed is related to the sampling frequency of the regulator. Depending on the desired accuracy and the situation of the regulator in PWM operation, the rotational speed threshold value can be determined such that the sampling frequency corresponds to at least two times up to twenty times the frequency corresponding to the rotational speed threshold value. The threshold rotational speed up to which the regulator operates steadily is also related to the speed of the process used.

Regulating the phase current until the pole rotor voltage is less than the limit of the intermediate circuit voltage (range I) or until a rotational speed threshold n_Gr(to the left of range III) and generate a pulse pattern via PWM. As soon as the pole rotor voltage is greater than the intermediate circuit voltage when the rotational speed threshold is reached, the phase current is no longer regulated, but rather a maximum phase voltage is set in the square wave pattern (blockmaster) and the advance commutation angle is regulated.

In the case shown in fig. 5a, the rotational speed threshold is now so large that the rotational speed threshold n is exceeded_GrPreviously, range II was always reached depending on the current torque. This means that a PWM-square wave transition always occurs without a voltage step. Square wave operation can also be used in range II.

In contrast, in the situation shown in fig. 5b it can be seen that: rotational speed threshold n for low torques_GrLess than the ideal transition point for square wave operation, i.e. at a location where the ranges I and III are mutually critical. In this case, the rotational speed threshold must be abruptly changed into the square-wave operation due to the stability of the phase current regulation.

If a transition from a PWM control to a square wave (Block) control is carried out at this rotational speed, a voltage step and a torque step are present.

This behavior is shown in fig. 6. For this purpose, the torque M, the rotational speed n, the current I and the voltage U are plotted with respect to the time t. At a point in time t₀A transition from PWM control to square wave (Block) control is carried out, with a rotational speed of approximately 3800/min.

The phase voltage U is visible at the transition point_dAnd U_qStep (2). The torque here being of the order of about-1.7 NmJump to +4Nm and phase current I_dAnd I_qOscillation is started. The regulation in the phase voltage is carried out until a complete step of approximately 15ms has elapsed. This problem is solved by the current method, which is set forth in more detail below.

Fig. 7, 8 and 9 illustrate three different actuation types that can be used in a preferred embodiment in the method according to the invention. The individual actuation types can be implemented as actuators or in the range of an actuator.

Fig. 7 shows a PWM control or PWM operation with field-oriented control. The measurement is provided with the reference character I_A...EAnd converted to I in space vector expression_qAnd I_d. In addition, the excitation current I in the excitation winding is detected_ex. The parameters P of the electric machine, i.e. for example the inductance and the flux linkage, are determined by means of the current I via a polynomial and taking into account the saturation behavior_q、I_d、I_exTo calculate. From the parameter P and the theoretical torque M_sollIn calculating the reference current I_q,refAnd I_d,refAnd is converted into a voltage reference value U by field-oriented regulation_dAnd U_q. For calculating the reference current, the temperature T of the stator and the resistance R associated therewith can be taken into account_S。

Subsequently, the reference value is converted into a pulse pattern 180 in the PWM controller, for example, the pulse pattern is center-aligned and set at the phase of the electric machine 100 via an inverter or a current transformer. Here, the rotor position angle can also be obtained via the sensor 170And electrical angular frequency omega_elAnd considered subsequently.

In fig. 8, a regulated square-wave control or square-wave operation with a leading commutation angle is shown. Instead of regulating the current, the parameter P of the electric machine and the current I are used as a slave_q、I_dCalculating the current torque M_ist. The current torque and the theoretical torque M_sollComparing and adjusting the leading commutation angle based on the deviation (preferably PI-regulator), the resulting leading commutation phase, referred to herein as αThe angle is directly converted into a square wave pattern 181, in which the largest phase voltage is always fed. Furthermore, reference is made to the description of FIG. 7, but wherein the electrical angular frequency ω is not required here_el。

Fig. 9 shows a transient control with q-value regulation of the phase voltage. This is a mixture of the two previously described types of manipulation. Again apply the current torque M_istAnd theoretical torque M_sollComparing, but adjusting or setting only the phase voltage U_dThe value of d of (a). Starting with an initial value U_q,initIncreasing the phase voltage U by a fixed gradient DeltaU_qQ value of (2). It is conceivable to consider the voltage limit U_Gr. The initial value is the last value of the square wave manipulation or the PWM manipulation. Subsequently, the pulse pattern 182 is set here by means of a PWM mechanism.

Fig. 10 shows a further preferred embodiment of the transition between the individual control types of the method according to the invention.

If pole rotor voltage U_pGreater than the intermediate circuit voltage U_dcMultiplied by a factor of 2/Pi, then there is PWM manipulation A₁And square wave manipulation A₃Conditions of direct transition between and without the need for transition manipulation a₂. The factor 2/Pi allows a conversion into a maximum settable sinusoidal natural oscillation in square wave operation.

If according to the rotation speed threshold value n_GrAnd a transition is made, then transition maneuver A is used₂. In each condition hysteresis or deviation O can be used, respectively, in order to avoid a persistent switching state.

If starting from PWM maneuver A₁Using transition manipulation A₂And satisfies the condition that the rotation speed n is greater than the rotation speed threshold n_Gr(n>n_Gr) ", then the q value of the phase voltage is increased with a fixed gradient until the condition" phase voltage U "is satisfied_dqIs greater than or at least approximately equal to the intermediate circuit voltage U_dcMultiplied by a factor of 2/Pi ". The value of d of the phase voltage is adjusted to the respective required torque. Physically U_qThe increase in d-current causes the field to be amplified, the pole rotor voltage to become larger and a frictionless transition to be achieved.

If the condition "n" is violated at the time of transition>n_Gr", then as long as this condition is invalid or violated, the q value of the phase voltage is no longer incremented, but rather decremented, and switched back to PWM operation.

Using transition maneuver A if starting from Square wave operation₂Then U is initialized with the last value in the square wave operation_qAnd decremented. Because in PWM control A₁The medium-voltage transition is not known in advance, so that the condition is that the set d current is smaller than the reference I in PWM operation_d,ref. This shows that the field is more greatly reduced than would be optimal in the respective operating point, and thus is a favorable transition condition.

In fig. 11, the variables in fig. 6 are re-illustrated, however, transition manipulation is used here. Instead of kicking up U_dAnd U_qNow slowly increments U_qE.g., a gradient of 400V/s. U shape_dApproximately remains constant because U_dIn this case, the current I is influenced_qTherefore I is_qThe set torque has a significant effect, and the required torque is approximately constant (adjusted to a constant power here) in the illustrated range. The d current increases from about-31A to nearly over 0A. In comparison with fig. 6, a constant torque is obtained from now on in the transition.