Method and apparatus for regulating an electric machine

Abstract

Method (100) for regulating an electric machine (310) with the steps: determining (110) a desired switching angle (u*); determining (120) a desired state (y*); determining (130) an actual state (y); determining (140) a difference (d) from the desired state (y*) and the actual state (y); determining (150), by means of a regulator (320), a switching angle adjustment (delta_u) from the difference (d); actuating (160) the electric machine (310) by means of a sum (u) from the desired switching angle (u*) and the switching angle adjustment (delta_u).

Description

Technical Field

[0001] This invention relates to a method and apparatus for regulating an electric motor. Furthermore, the invention also relates to a transmission system having corresponding equipment, a vehicle having a transmission system, a computer program, and a machine-readable storage medium. Background Technology

[0002] In the field of motor vehicle drive technology, for example, it is known from DE 10 2010 061897 A1 that electric motors are used as drive devices for vehicles. In such electric vehicles, the electric motor is used as a drive motor. To control the electric motor in the motor vehicle, power electronics are used, which include a reverse commutator that converts the DC voltage / DC current of the vehicle's onboard (high-voltage) battery into AC current. The reverse commutator typically has multiple energizable power switches. With the aid of a control device, the power switches are controlled in a pulse-width modulation manner so that the motor produces a defined torque at a defined speed of the motor's power output shaft in engine operating mode. To control such a reverse commutator for the motor, it is known to use field-oriented regulation (also known as vector regulation). In this case, a spatial vector (e.g., a current vector) moves, which rotates with the motor's power output shaft. In other words, the phase current required to control the motor is converted into a coordinate system (the so-called dq system) where the rotor is fixed and rotates with the machine's magnetic field. In field-oriented regulation, instead of the phase current (AC variable), the current components Id and Iq, transformed in this way, are regulated as DC variables, and the desired voltage to be set on the machine is calculated. Since the inverter can only exhibit discrete, pulsating voltage changes, the continuous voltage must be converted to the switching mode of the power electronics. A modulator, located downstream of the regulator, undertakes this task. The modulator is responsible for applying the correct voltage on average during one switching cycle (Schaltzyklus) of the electronics. This switching cycle is significantly shorter than the electrical cycle of the voltage to be set. Because of this, a dynamic response to changes in reference variables (such as the desired current or torque) can be achieved in the regulator system. The switching pulse or switching action time is calculated here, for example, by comparison with trigonometric functions (Sine-Triangle PWM) or through simple trigonometric calculations (Space Vector PWM). This limits the possibility of switching in a loss-optimized manner.

[0003] Another method for achieving highly dynamic regulation is the hysteresis-based direct-switching regulator. Here, a continuous (or high-frequency scan) comparison is made: whether a reference variable, such as current, is within a tolerance band. If a violation of this band occurs, the power electronics are directly switched, depending on the type of violation. While this method results in very dynamic following of the reference variable, such as current or torque, it also leads to a nearly uncontrollable spectrum of the reference variable (including subharmonics), an uncertain switching frequency, and trending high losses. Therefore, it is uncommon in practical use. Thus, field-oriented regulation can achieve a highly dynamic response to changes in the reference variable, such as speed or torque, and can achieve the lowest possible losses.

[0004] Here, due to harmonics or the switching of electronic devices, losses can be monitored either entirely or conditionally. In contrast, by applying optimized pulse patterns to power electronics, accurate loss setting can be consistently achieved. Here, the machine is operated using an offline pre-optimized pulse pattern. A pulse pattern is a sequence of on / off states of the power semiconductor during one electrical cycle, defined or derived from the on and off times or switching angles of the power semiconductor during that electrical cycle. The switching pattern is optimized with respect to an arbitrary cost function during one electrical cycle. Loss optimization can be achieved using pulse patterns if the cost function is chosen such that it characterizes the weighted loss. Compared to patterns generated using PWM, the voltage is correctly set only during one electrical cycle and not through short scan steps. For this reason, in transient situations, unwanted voltage patterns occur on the machine when the pattern changes abruptly (or the pattern sequence is rapidly traversed). These voltage patterns cause a severely distorted volt-second balance in the coil, and thus generally lead to excessive overshoot. This can not only cause persistent damage to power electronics, but also result in long set times. Therefore, this method is not practically applicable. Consequently, regulation systems based on this pattern exhibit significant drawbacks in terms of dynamics or strong overshoot with respect to the reference variable. Exemplarily for a direct-switching and highly dynamic method based on optimal fixed pulse patterns, the publication "Model predictive pulse pattern control" by GEYER, Tobias et al. (IEEE Transactions on Industry Applications, 2011, Vol. 48, No. 2, pp. 663-676) shows the online adjustment of offline-generated switching patterns to meet dynamic requirements. The optimization problem is solved online to adjust the switching angle (of the power semiconductor with respect to the electrical fundamental wave). Here, the deviation from the variables calculated offline is considered in the optimization function. Trajectories of the stator flux for the ASM are calculated from the switching patterns, and these trajectories should be tracked in an regulated manner with as little deviation as possible. For online optimization, a strongly simplified model based on the stator flux equation of the ASM is used here, neglecting resistance. The optimization function here penalizes deviations from the variables calculated offline, so as to achieve optimality in the quiescent state. The result of the optimization is the following switching timing: the switching timing is directly fed to the power electronics by the regulator.

[0005] There is a need for a method that combines the advantages of two approaches: magnetic field-oriented regulation and offline pre-optimized pulse or switching modes. Summary of the Invention

[0006] A method for adjusting a motor is provided, the method comprising the steps of:

[0007] Determine the desired switching angle;

[0008] Determine the desired state;

[0009] Determine the actual state;

[0010] Determine the difference between the desired state and the actual state;

[0011] The switching angle adjustment is determined based on the difference using a regulator;

[0012] The motor is controlled by the sum of the desired switching angle and the switching angle adjustment.

[0013] A method for regulating a motor is provided. To determine at least one desired switching angle, preferably using the most accurate model of the machine as possible, and based on a cost function, for example, describing different weighted loss terms of the machine, the optimal pulse pattern (OPP) is calculated offline. The result of this optimization is the desired switching moment or desired switching angle during one electrical cycle. This explicitly characterizes the change process of the desired switching angle. Preferably, the desired switching change process is read from the current rotor position or rotor angle. The terms "desired switching moment (0...1)" or "desired switching angle (0...360°)" are considered to have the same meaning in the following text, as they are both related to one electrical cycle and can therefore be explicitly converted to each other. Preferably, the desired switching angle is derived from the product of the angular velocity and the desired switching moment. To determine at least one desired state, the determined desired switching angle is preferably given to a model G of the segment, which in turn represents the characteristics of the real segment as accurately as possible, preferably the characteristics of the drive unit or transmission system, in order to obtain the desired state (i.e., current or magnetic flux). Preferably, in a suitable reference frame (dq, The desired state is generated from (a, b, c). Therefore, the method is not limited to using a defined reference variable. The calculation of the desired state is preferably performed offline so as not to burden the computational power required by the control unit. Preferably, the desired state is read from the results of the offline calculation stored in the form of a lookup table. To determine at least one actual state, the state variables of the actual segment are preferably determined. The difference between the desired state and the actual state is determined by subtraction. To determine the switching angle adjustment based on the difference by means of a regulator, the difference between the desired state and the actual state is preferably fed to the regulator, preferably to a model prediction regulator. Based on the determined switching angle adjustment, the motor is controlled, preferably by means of the sum of the desired switching angle and the switching angle adjustment, preferably by means of an inverter.

[0014] Repeatedly, preferably interleaved in time, determining the desired switching angle leads to the determination of the process of change of the desired switching angle. This means that, in order to determine the process of change of the desired switching angle or the trajectory of the desired switching angle, this method is preferably repeated. The corresponding content applies to determining the desired state or the actual state. Preferably, repeatedly, preferably interleaved in time, determining the desired state or the actual state leads to the determination of the process of change of the desired state or the process of change of the actual state, or the trajectory of the desired state or the trajectory of the actual state. The corresponding content applies to determining the adjustment of the switching angle. Preferably, repeatedly, preferably interleaved in time, determining the adjustment of the switching angle leads to the determination of the process of change of the switching angle adjustment or the trajectory of the switching angle adjustment.

[0015] Therefore, a method is provided for regulating a motor to achieve highly dynamic operation of a drive system and optimal operation under static conditions. Static optimality is established by considering the difference between the desired and actual states, i.e., the Delta expression of the system, in relation to any previously calculated criteria, such as losses or intermediate circuit fluctuations. Therefore, the regulator preferably does not observe absolute variables (such as flux, current, etc.), but only the reference value or the deviation from the reference change process (i.e., the delta variable). The regulator simply adjusts the delta variable to zero, thereby minimizing these deviations to values ​​optimized offline. This significantly reduces the overhead of online calculations by the regulator and simultaneously improves performance. Thus, preferably, under nominal conditions, the regulation is highly dynamic and optimal under static conditions. This method directly predefines the switching timing of the power electronics, i.e., eliminating the need for an intermediate modulator. This method can operate with different reference variables (current or flux). The dynamic model upon which the regulator's operation and design are based is a model in Delta or error coordinates. This model merely describes the difference between the desired and actual states, or the difference between the desired and actual state trajectories (deviations from the trajectories), in order to simulate and adjust errors in advance in a time-discrete or angle-discrete manner. The application of this approach to Delta regulators is improved through structural adjustments and also adjustments to the modeling techniques. Using a universally valid Delta expression, this scheme can also be applied using current instead of flux as the reference variable. Furthermore, any model can be used. This is advantageous when idealized assumptions must be discarded, for example in the case of a coupled 6-phase PSM, but also for a wide range of other machines. With a universally valid expression in error coordinates, preferably in the case of a predictive regulator, the deviation of the controlled variable from the reference variable is included in each individual prediction step of the model predictive regulator and penalized in the optimization. This improves the possibility of targeted regulator configuration and reduces the correlation between regulator dynamics and the prediction time domain. Furthermore, time-varying systems are significantly better tuned through stepwise linearization, since stepwise linearization is performed in the time domain.

[0016] Advantageously, a method combining the advantages of both approaches is provided in which the offline-optimized impulse patterns or switching moments are dynamically adjusted online via model-predicted regulation. This is achieved through a simple model and a short time domain representation in the Delta / error coordinates. The regulation concept is optimized with static loss and, despite this, highly dynamic.

[0017] For the entire electric drive system, loss optimality is advantageously derived in the sense of cost function. The characteristics of this application (e.g., transmission noise) are considered in offline optimization without the influence of adjusted dynamic losses. Closed-loop performance is unaffected by simplifications during modeling, ensuring that even high tolerances in manufacturing do not necessarily compromise closed-loop performance from an adjustment perspective.

[0018] In another embodiment of the invention, the actual state characterizes the actual phase current through the motor, while the desired state characterizes the desired phase current through the motor.

[0019] The actual phase current through the motor is preferably characterized by the actual state, which is preferably determined by a suitable measuring device; and the desired phase current is preferably characterized by the desired state, which is preferably determined based on a pre-given torque. Advantageously, variables of a specific variant that enable regulation of the motor are assigned to the state or trajectory.

[0020] In another embodiment of the invention, at least one desired state is stored in a family of characteristic curves based on parameters torque, speed, and rotor angle. For this adjustment, at least one desired state is determined from the family of characteristic curves based on the parameters torque, speed, and rotor angle.

[0021] The desired state is stored in a family of characteristic curves based on multiple parameters, and is determined from this family of characteristic curves based on these multiple parameters using this method. Advantageously, a method is provided that can consider multiple objectives or dimensions.

[0022] In another embodiment of the invention, at least one desired state is determined by means of a machine model, wherein the machine model takes into account the inductance matrix, resistance matrix, magnetic flux and / or angular velocity of the motor.

[0023] Advantageously, a method for considering machine models is provided.

[0024] In another embodiment of the invention, the optimization problem is solved using a regulator. Preferably, the regulator is a model prediction regulator.

[0025] Optimization problems are solved using regulators, preferably convex quadratic optimization problems (QP). Advantageously, a method using regulators is provided for efficient multivariable regulation of motors.

[0026] Furthermore, the present invention also relates to a computer program comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method currently described.

[0027] Furthermore, the present invention also relates to a computer-readable storage medium comprising instructions that, when executed by a computer, cause the computer to perform the steps of the method currently described.

[0028] Furthermore, the present invention also relates to a device for regulating a motor, the device having an regulating mechanism. The regulating mechanism is configured to determine a desired switching angle, a desired state, an actual state, determine a difference between the desired state and the actual state, determine a switching angle adjustment based on the difference using a regulator, and control the motor using the sum of the desired switching angle and the switching angle adjustment.

[0029] An apparatus is provided, comprising an adjustment device including a regulator and preferably a model. The adjustment device is configured to determine a desired switching angle, preferably using the most accurate possible model of the section to be adjusted and based on a cost function. The adjustment device is further configured to determine a desired state based on the determined desired switching angle. The adjustment device is further configured to determine an actual state. The adjustment device is configured to determine the difference between the desired state and the actual state by means of subtraction. The adjustment device is configured to determine a switching angle adjustment based on the difference using the regulator. The adjustment device is further configured to control a motor based on the determined switching angle adjustment, preferably using an inverter to control the motor.

[0030] Advantageously, a device for effectively regulating a motor is provided.

[0031] Furthermore, the present invention also relates to a drivetrain incorporating a motor and the described device. For example, such an electric drivetrain is used to drive electric vehicles. With the aid of the described method and device, optimized operation of the drivetrain can be achieved.

[0032] Furthermore, the present invention also relates to a vehicle having the described drivetrain. Advantageously, a vehicle is therefore provided that includes means for effectively regulating the motor.

[0033] It should be understood that the features, characteristics, and advantages of the method according to the invention are correspondingly suitable for or applicable to the device or the transmission system and the vehicle, and vice versa. Attached Figure Description

[0034] Other features and advantages of the invention will become apparent from the following description with reference to the accompanying drawings.

[0035] Figure 1 A schematic diagram of a regulator structure for a method of regulating a motor is shown.

[0036] Figure 2 A vehicle with a drivetrain is shown schematically.

[0037] Figure 3 A flowchart illustrating a method for regulating a motor is shown. Detailed Implementation

[0038] Figure 1 A regulator structure 300, namely a Delta regulator, for a method of regulating motor 310 is shown, which has an regulating device 340. The regulating device 340 includes a regulator 320 and preferably includes a model 330, a difference point 322, and / or a summation point 324. A desired switching angle u* is determined and is fed to the regulator 320 and the model 330, preferably to a physical model of the section to be regulated or to a machine model. Using this model, a desired state y* is determined based on the determined desired switching angle u*. The actual state y is further determined, preferably on the actual section to be regulated using a suitable measuring device. The difference d formed by the desired state y* and the actual state y is determined by subtraction at the difference point 322, and this difference d is fed to the regulator 320. The regulator 320 determines the switching angle adjustment delta_u based on the difference d, wherein the regulator 320 preferably also considers the desired switching angle u*. The determination of the switching angle adjustment delta_u is preferably performed in a model prediction manner. The desired switching angle change process u* is preferably considered here to avoid switching angle adjustments delta_u that result in unacceptable switching modes (e.g., due to the required dead time or minimum turn-on and turn-off times of the power semiconductor). Preferably, regulator 320 considers such conditions as auxiliary conditions. With the help of regulator 320, a direct-switching regulation concept is preferably developed, which calculates the switching angle adjustment delta_u or the adjustment of the switching timing based on the deviation d between the optimized desired state y* and the actual state y (preferably a measurement of variables on a real machine, preferably a permanent magnet synchronous machine PSM), and thus adjusts the existing deviation. This occurs by solving the optimization problem according to the "receding horizon" principle, i.e., involving a model prediction (MPC) scheme. The model or scheme preferably used in regulator 320 is not intended to perform a forward simulation of the entire system in the prediction time domain, but only to perform a forward simulation of the difference d. This method, which uses the difference d between the desired state y* and the actual state y, i.e., the expression in the "Delta / error coordinates", has a great advantage in adjustment because it is significantly simpler than the complete model of the motor.

[0039] When this scheme is preferably applied to a six-phase PSM with a 2×3 star connection, the creation and modeling of the regulator for the current used as a reference variable yields: y=x=iabc. Regulation is then used based on the following instantaneous deviation of the current on the actual machine and the reference trajectory: As in the general representation, the correction of switching moments is described using the switching angle adjustment delta_u (these switching moments can be explicitly mapped to the switching angle, and vice versa). Preferably, for the dynamics of the phase current, a time-continuous state model is derived:

[0040] .

[0041] In this equation, Labc describes the entire occupied 6×6 inductance matrix of the machine, R describes the (stator) resistance matrix, ψpm,abc describes the permanent magnet flux of the machine, and ωel describes the electric angular velocity with its corresponding angle φel. Matrix S contains a combination of matrices: the use of these matrices allows the use of terminal voltage uuvw instead of phase voltage uabc in the model. These two voltages are distinguished only by their reference potentials and apply uuvw = uabc + u*, where u* describes the neutral point potential of the motor. Subsequently, the corresponding re-expression is implemented in discrete coordinates. Furthermore, all matrices except for the resistances are time-varying. Using the model described above, an optimized pulse pattern or voltage variation process is calculated, from which the current trajectory is calculated through forward simulation and scanning (assuming a stationary state). The calculated pulse pattern and the scanned trajectory are stored as a family of characteristic curves with respect to speed and torque, and can be loaded and used according to the machine's operating state. Therefore, online generation of the reference or desired state y* is not necessary, which minimizes computational overhead. The calculated voltage and current variations over one electrical cycle result in a pulsating voltage pattern that follows a sinusoidal fundamental wave and exhibits prominent harmonic components. This pulsating voltage pattern corresponds to optimal machine operation with losses. Therefore, the current variation should be tracked during regulated operation. This is achieved by optimally offsetting the voltage edges using regulator 320. This is then described in the Δ / error coordinate system:

[0042] ,

[0043] The system is derived as follows:

[0044] .

[0045] Here, It describes the instantaneous deviation between the current and the desired state trajectory or reference trajectory on a real machine, and This describes the applied voltage difference compared to the optimized pulse mode. It is clear that all terms independent of iabc(t) and uuvw(t) are neglected, thus reducing the parameter dependence of the regulation. Through time discretization, the following model is derived:

[0046] .

[0047] The integral of the input voltage over the scan times tk and tk+1 can be solved to obtain the desired expression as a linear model of the switching action time (or its offset or switching angle):

[0048] .

[0049] Through phase-by-phase observation, for phase p, this leads to the following expression:

[0050]

[0051] vector This includes voltage. and The difference before switching process j or j+1, and applicable , where np describes the number of switching processes in phase p during the time interval [tk, tk+1]. Variables The switching time j in phase p is described. Therefore, The reference switching time for phase p is described. and the actual switching moment The difference vector. By applying this relation to all phases and generalizing in matrix form, we obtain:

[0052] .

[0053] It is clear that, with Figure 1 Compared to the diagram in the image, the settings are... and During the scan period Ts, the machine's current speed / angle and current are measured. Based on the pre-given speed / angle and torque, a suitable, previously calculated optimal pulse pattern is then read from a stored family of characteristic curves. Specifically, the reference current and reference switching time are extracted. The deviation between the current and the reference value is calculated, and this deviation, along with the switching angle sequence to be modified, is fed to the MPC regulator or regulator 320. In the regulator, a suitable adjustment for the switching time or the switching angle adjustment delta_u is determined based on an optimization problem.

[0054] Summation at summation point 324 forms sum u, which is the desired switching angle u* and the switching angle adjustment delta_u. Summ u includes the switching sequence to be applied. With the aid of sum u, motor 310 is controlled, which preferably includes an inverter 312 and an electric drive unit 314.

[0055] The previous explanation implicitly started from a measurable state, that is, y=x. If this is not the case, then the observer must be expressed. This is the case in the described application if flux instead of current is used as the reference variable. The observer can also be expressed using error coordinates, and such expression is also advantageous.

[0056] The optimization problem is solved using regulator 320, and the optimization problem is as follows:

[0057]

[0058] Index p describes the individual phases of the machine and takes a value between 1 and 3 times the number of systems. For a preferred six-phase machine with a 2×3 connection, Nsys=2 is applicable because there are two three-phase systems. The norm equipped with indices Q and R is... or The squared weighted 2-norm. Here, Q and R are commonly used matrices in MPC regulator design. The last two auxiliary conditions, expressed as equations, characterize the error dynamics that have been verified for reliability above. Preferably, other auxiliary conditions should be considered, preferably other auxiliary conditions for state constraints. For the application described above, The current must be set up as a reference variable, and the matrices Ar, Br (indicated by index k) and Cr=I, dr=0 derived above must be used. This optimization problem is a convex quadratic optimization problem (QP). This optimization problem can be solved using standard methods, such as the "active set" or "interior point" methods.

[0059] Figure 2 A vehicle 500 with a drivetrain 400 is shown schematically. The drivetrain 400 includes an electric motor (310) and a device with an adjustment mechanism 340. The figure exemplarily shows a vehicle with four wheels, wherein the invention can also be used in any vehicle with any number of wheels, whether on land, water, or in the air.

[0060] Figure 3A schematic flow diagram of a method 100 for adjusting motor 310 is shown. The method begins at step 105. In step 110, a desired switching angle u* is determined. In step 120, a desired state y* is determined. In step 130, the actual state y is determined. In step 140, a difference d is determined from the desired state y* and the actual state y. In step 150, a switching angle adjustment delta_u is determined based on the difference d using a regulator 320. In step 160, motor 310 is operated using the sum u formed by the desired switching angle u* and the switching angle adjustment delta_u. The method ends at step 195.

Claims

1. A method (100) for regulating an electrical machine (310), having the steps of: determining (110) a desired switching angle (u*); determining (120) a desired state (y*); determining (130) an actual state (y); determining (140) a difference (d) made up of the desired state (y*) and the actual state (y); determining (150), by means of a regulator (320), a switching angle adjustment (delta_u) from the difference (d); actuating (160) the electrical machine (310) by means of a sum (u) made up of the desired switching angle (u*) and the switching angle adjustment (delta_u), wherein the actual state (y) characterizes an actual phase current (i_a,b,c) through the electrical machine, and the desired state (y*) characterizes a desired phase current (i*_a,b,c) through the electrical machine, wherein the switching angle adjustment (delta_u) is determined on the basis of an optimization problem, wherein the optimization problem is a convex quadratic optimization problem (QP). The desired state (y*) is deposited in a family of characteristic curves as a function of a parameter torque, a parameter rotational speed and a parameter rotor angle, and the desired state (y*) is determined from the family of characteristic curves for the regulation as a function of the parameter torque, the parameter rotational speed and the parameter rotor angle. The desired state (y*) is determined by means of a machine model (330), wherein the machine model (330) takes into account an inductance matrix, a resistance matrix, a magnetic flux and / or an angular velocity of the electrical machine. The optimization problem is solved by means of the regulator (320).

5. A computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps / methods of the method (100) according to any one of claims 1 to 4.

6. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the steps / methods of the method (100) according to any one of claims 1 to 4.

7. A device (300) for regulating an electrical machine (310), having a regulating apparatus (340), wherein the regulating apparatus (340) comprises a regulator (320), and wherein the regulating apparatus (340) is set up to: determine a desired switching angle (u*); determine a desired state (y*); determine an actual state (y); determine a difference (d) made up of the desired state (y*) and the actual state (y); determine, by means of a regulator (320), a switching angle adjustment (delta_u) from the difference (d); actuate the electrical machine (310) by means of a sum (u) made up of the desired switching angle (u*) and the switching angle adjustment (delta_u), wherein the actual state (y) characterizes an actual phase current (i_a,b,c) through the electrical machine, and the desired state (y*) characterizes a desired phase current (i*_a,b,c) through the electrical machine, wherein the switching angle adjustment (delta_u) is determined on the basis of an optimization problem, wherein the optimization problem is a convex quadratic optimization problem (QP). The desired state (y*) is deposited in a family of characteristic curves as a function of a parameter torque, a parameter rotational speed and a parameter rotor angle, and the desired state (y*) is determined from the family of characteristic curves for the regulation as a function of the parameter torque, the parameter rotational speed and the parameter rotor angle.

2. The method for regulating an electric machine according to claim 1, wherein, The desired state (y*) is determined by means of a machine model (330), wherein the machine model (330) takes into account an inductance matrix, a resistance matrix, a magnetic flux and / or an angular velocity of the electrical machine.

3. The method for regulating an electric machine according to claim 2, wherein, The optimization problem is solved by means of the regulator (320).

4. The method for regulating an electric machine according to any of the preceding claims, wherein, ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 8. A drive train (400) having an electric machine (310) and an apparatus (300) according to claim 7.

9. A vehicle (500) having a drive train (400) according to claim 8.

Images