METHOD FOR OPERATING AN ELECTRIC MACHINE, DEVICE FOR OPERATING AN ELECTRIC MACHINE, ELECTRIC DRIVE SYSTEM

Description

[0001] The invention relates to a method for operating an electrical machine having a rotatably mounted rotor and a motor winding, wherein the motor winding is electrically connected to an electrical energy storage device by power electronics, wherein the machine and / or a device having the machine are monitored for faults, and wherein the motor winding is short-circuited by controlling the power electronics when a fault is detected.

[0002] Furthermore, the invention relates to a device for operating an electrical machine, comprising a control unit.

[0003] Furthermore, the invention relates to an electric drive system. State of the art

[0004] The publications CN 107 294 450 A and US 2008 / 116842 A1 disclose methods and devices for operating an electrical machine. An electrical machine typically has a rotatably mounted rotor and a motor winding. The rotor can be driven or rotated by appropriately applying current to phases of the motor winding. For example, the motor winding is a stator winding fixed to the housing and arranged distributed around the rotor. In order to achieve the desired current supply to the motor winding or the phases of the motor winding, the motor winding is typically electrically connected to an electrical energy storage device by power electronics. For example, the power electronics has a different half-bridge with two switching elements for each phase of the motor winding. The current supply to the motor winding is then regulated or controlled by the control of the power electronics.

[0005] It is known from the prior art to transfer the electrical machine to a safe state in the event of a fault. The machine and / or a device comprising the machine are therefore monitored for faults, and the machine is transferred to the safe state when a fault is detected. For example, an active short circuit (AKS) of the motor winding is set as the safe state. In this respect, when a fault is detected, the motor winding is short-circuited by controlling the power electronics. This is achieved, for example, by controlling the switching elements of the power electronics in such a way that the switching elements switch into conduction. The faults can be faults in the electrical machine itself or faults in the device comprising the electrical machine.For example, if the machine is part of a motor vehicle, the faults may also be faults of the motor vehicle. Possible faults include sensor errors, software errors, overvoltage or overcurrent errors, defects or partial defects in drivetrain components, or an accident involving the motor vehicle.

[0006] Various procedures for short-circuiting the motor winding are known. In the so-called hard short-circuiting, the motor winding is actively short-circuited from the current actual operating point. In the hard short-circuiting, a transient short-circuit current occurs in the motor winding, which can be so high that it demagnetizes the permanent magnets of the rotor. To reduce the transient short-circuit current, a soft short-circuiting can be carried out. In this case, a voltage vector is sequentially shortened until the zero voltage, i.e. the active short circuit, is reached. Such a method is described, for example, in the published patent application WO 2015 090 754 A1. From the published patent application EP 3 545 617 A1, a method for short-circuiting a motor winding is known, which is carried out as a function of an intermediate circuit voltage of an intermediate circuit capacitor of the power electronics. Disclosure of the invention

[0007] The method according to the invention is characterized by the features of claim 1 in that a target trajectory for an actual current vector of an electrical motor current flowing through the motor winding is determined, wherein the target trajectory runs from a current actual operating point of the machine to a short-circuit operating point of the machine, wherein a feedforward control is predicted as a function of the target trajectory, and wherein the power electronics for short-circuiting the motor winding are controlled as a function of the feedforward control in such a way that a profile of the actual current vector when short-circuiting the motor winding at least substantially corresponds to the target trajectory. The motor current is understood to be the totality of the electrical phase currents flowing through the phases of the motor winding. In this respect, the current vector describes the electrical phase currents.The current vector preferably describes the phase currents relative to a rotor-fixed coordinate system. A torque-generating current iq forms a first directional component of the current vector. A flux-generating current id forms a second directional component of the current vector. The short-circuit operating point of the electric machine is the operating point that the machine assumes when the motor winding is short-circuited. Each operating point of the electric machine corresponds to a different current vector. According to the invention, a target trajectory is determined that runs from the current actual operating point to the short-circuit operating point.Depending on the target trajectory, a feedforward control is predicted and the power electronics are controlled in accordance with the feedforward control such that the course of the actual current vector when short-circuiting the motor winding at least essentially corresponds to the target trajectory. By determining the target trajectory and correspondingly controlling the power electronics, a desired transition of the electrical machine from the current actual operating point to the short-circuit operating point can be specified particularly precisely. For example, it can be achieved that the short-circuit operating point is set in a time-optimized manner and with a low transient short-circuit current. Preferably, the target trajectory is determined in such a way that the transition from the actual operating point to the short-circuit operating point is at least essentially time-optimized.Preferably, a control sequence comprising multiple target voltage vectors is predicted as the feedforward control. To control the power electronics as a function of the feedforward control, multiple control signals for the power electronics are determined as a function of the control sequence, and the power electronics are controlled as a function of the control signals.

[0008] According to a preferred embodiment, the target trajectory is determined as a function of the current actual operating point during operation of the electric machine. The feedforward control is thus determined "online" as a function of the current actual operating point. This results in the advantage that a suitable target trajectory can be determined for any current actual operating point. Accordingly, based on any current actual operating point, the short-circuit operating point can be set along the target trajectory. If the target trajectory is determined during operation of the electric machine, the feedforward control is also predicted during operation of the electric machine.Preferably, the feedforward control predicted during operation of the electric machine is temporarily stored for a predetermined period of time, so that the predicted feedforward control is available for the predetermined period of time and can be used as the basis for controlling the power electronics.

[0009] The target trajectory is preferably determined in preliminary tests. The target trajectory is thus determined "offline," for example, during the calibration of the electric machine in the factory. Accordingly, the target trajectory does not need to be determined during operation of the electric machine. This results in the advantage that no computing capacity needs to be reserved for determining the target trajectory in a control unit designed to determine the control signals for the power electronics. Another advantage is that the feedforward control is available particularly quickly compared to determining the target trajectory during operation of the electric machine. If the target trajectory is determined in preliminary tests, the feedforward control is preferably also predicted in the preliminary tests. The feedforward control is then stored in a data memory assigned to the electric machine.Alternatively, the control signals are preferably also determined in the preliminary tests, in which case the control signals are then stored in the data memory.

[0010] Preferably, a target trajectory is determined for each of several potential actual operating points. This means that several potential actual operating points are specified, and a target trajectory is determined for each of these actual operating points. This is particularly advantageous if the target trajectories are determined in preliminary tests. For example, several potential actual operating points are specified, each of which lies on an MTPA (Maximum Torque Per Ampere) characteristic curve of the electric machine, and a target trajectory is determined for each of these actual operating points. If a target trajectory is determined for each of several actual operating points, a feedforward control is preferably also predicted for each of the target trajectories. These multiple feedforward controls are then stored in the data memory, particularly preferably in a characteristic map.Alternatively, the control signals determined as a function of the multiple pre-controls are preferably stored in the characteristic map.

[0011] Preferably, the target trajectory is determined based on a model of the electrical machine. The inherent dynamics of the electrical machine are therefore taken into account when determining the target trajectory. The model of the electrical machine describes, in particular, the relationship between the electrical terminal voltages applied to the phases of the motor winding, on the one hand, and the phase currents caused by the terminal voltages, on the other. For example, the model contains information regarding the inductances and electrical resistances of the machine.

[0012] According to a preferred embodiment, a threshold current value is specified, wherein the target trajectory is determined as a function of the threshold current value such that a current value of the actual current vector always falls below the threshold current value when short-circuiting the motor winding. Current values ​​that exceed the threshold current value are thus avoided. Such high current values ​​could lead to demagnetization of the rotor's permanent magnets and are therefore undesirable.

[0013] Preferably, a threshold voltage value is specified, with the target trajectory being determined as a function of the threshold voltage value such that the voltage values ​​of the electrical terminal voltages of the machine always fall below the threshold voltage value when the motor winding is short-circuited. This also prevents overloading of the electrical machine during short-circuiting.

[0014] According to a preferred embodiment, the target trajectory is determined as a function of an intermediate circuit voltage of an intermediate circuit of the power electronics. By taking the intermediate circuit voltage into account, the intermediate circuit or an intermediate circuit capacitor of the intermediate circuit can be protected from overvoltages when the motor winding is short-circuited. In particular, an electrical connection between the energy storage device and the power electronics is interrupted during short-circuiting. This is referred to as load shedding. The energy storage device then no longer has a charge-balancing effect on the intermediate circuit. By taking the intermediate circuit voltage into account when determining the target trajectory, the intermediate circuit capacitor or the intermediate circuit can still be protected from overvoltages.

[0015] Preferably, a voltage threshold is specified for the intermediate circuit voltage, with the target trajectory being determined as a function of the voltage threshold in such a way that the intermediate circuit voltage always falls below the voltage threshold when the motor winding is short-circuited. This prevents overvoltages exceeding the voltage threshold.

[0016] According to a preferred embodiment, it is provided that a target voltage for the intermediate circuit voltage is specified, wherein the target trajectory is determined as a function of the target voltage such that the intermediate circuit voltage corresponds to the target voltage when the short-circuit operating point is reached. Expediently, an electrical voltage is specified as the target voltage which does not damage the intermediate circuit. For example, the nominal voltage is the rated voltage of the intermediate circuit capacitor. So that the intermediate circuit voltage corresponds to the target voltage when the short-circuit operating point is reached, the target trajectory is determined or selected in particular such that the intermediate circuit is at least temporarily discharged when the motor winding is short-circuited.

[0017] According to a preferred embodiment, the target trajectory is determined by a model-predictive controller. Model-predictive control is generally known from the prior art and is also referred to as model predictive control (MPC). The model-predictive controller can precisely determine a target trajectory that describes a time-optimized transition from the actual operating point to the short-circuit operating point, so that the feedforward control predicted as a function of the target trajectory achieves an at least essentially time-optimized setting of the short-circuit operating point. The model-predictive controller preferably determines the target trajectory as a function of the machine model, the specified threshold current value, and the specified threshold voltage value.Particularly preferably, the model-predictive controller also takes the intermediate circuit voltage of the intermediate circuit into account when determining the target trajectory, for example in the form of the voltage threshold and / or the target voltage. Preferably, the feedforward control is predicted by the model-predictive controller.

[0018] According to a preferred embodiment, it is provided that a sensor signal from at least one sensor is compared with a predetermined limit value, and that the fault is detected depending on the comparison. For example, it is detected that the fault has occurred if the sensor signal exceeds the limit value or if the sensor signal falls below the limit value. The comparison is expediently based on the sensor signal of a sensor in which at least one of the potential fault cases is evident when the fault occurs. For example, the fault is a defect or partial defect in a component of the electrical machine. Such defects cause, for example, overcurrents and / or overvoltages in the power electronics and / or the motor winding.Accordingly, the comparison is preferably based on the sensor signal of a sensor designed to monitor the power electronics and / or the motor winding. However, the fault may also be an accident involving the motor vehicle. Accordingly, the comparison is preferably based on the sensor signal of an acceleration sensor mounted on the body. Preferably, the sensor signals from different sensors, each with different limit values, are compared.

[0019] According to a preferred embodiment, when a fault is detected, the electrical machine is controlled to the short-circuit operating point depending on the feedforward control. The motor winding is therefore short-circuited in an uncontrolled manner depending on the feedforward control. Adapting the short-circuiting to a sensor signal from a control device is preferably omitted. A control device is understood to be a device whose sensor signal is taken into account in the control of the machine during normal operation of the machine. Examples of such control devices include an angle of rotation sensor assigned to the rotor, a current sensor assigned to the power electronics or the motor winding, and a voltage sensor assigned to the motor winding.This results in the advantage that the course of the actual current vector during short-circuiting corresponds to the desired target trajectory even if the fault affects one or more of the control devices.

[0020] The device according to the invention for operating an electrical machine having a rotatably mounted rotor and a motor winding, wherein the motor winding is electrically connected to an electrical energy store by power electronics, is characterized with the features of claim 14 by a control unit that is specifically designed to carry out the method according to the invention when used as intended. If the control unit is used as intended, the method according to the invention is carried out in the control unit or by the control unit. This also results in the advantages already mentioned. Further preferred features and combinations of features emerge from the description and the claims. The control unit is preferably designed as a microcontroller.

[0021] The electric drive system according to the invention comprises an electric machine having a rotatably mounted rotor and a motor winding, wherein the motor winding is electrically connected to an electrical energy storage device via power electronics. The drive system is characterized by the features of claim 15 and the device according to the invention for operating the electric machine. This also results in the aforementioned advantages. Further preferred features and combinations of features emerge from the description and the claims.

[0022] The invention is explained in more detail below with reference to the drawings. Figure 1 shows an electric drive system, Figure 2 shows a method for operating an electric machine of the drive system, Figure 3 shows a first target trajectory, Figure 4 shows current and voltage curves when short-circuiting a motor winding of the electric machine according to the first target trajectory, Figure 5 shows a second target trajectory and Figure 6 shows a current curve and a voltage curve when short-circuiting the motor winding according to the second target trajectory.

[0023] Figure 1 shows a schematic representation of an electric drive system 1 of a motor vehicle not shown in detail.

[0024] The drive system 1 comprises an electric machine 2. The electric machine 2 has a rotatably mounted rotor. Furthermore, the electric machine 2 has a stator winding as the motor winding. The stator winding is distributed around the rotor in such a way that the rotor can be rotated by appropriately applying current to the stator winding. In this case, the stator winding has three phases.

[0025] The drive system 1 also includes power electronics 3 with multiple switching elements. For example, the power electronics 3 includes a number of half-bridges corresponding to the number of phases, with each of the half-bridges having two switching elements. The power electronics 3 also includes an electrical intermediate circuit 5 with an intermediate circuit capacitor 6.

[0026] The stator winding is electrically connected to an electrical energy storage device 4 of the drive system 1 via the power electronics 3.

[0027] Two battery contactors 7 and 8 are assigned to the energy storage device 4. The battery contactors 7 and 8 can be used to establish or interrupt an electrical connection between the power electronics 3 and the energy storage device 4.

[0028] The drive system 1 also has a device 9 with a control unit 10. In this case, the control unit 10 is a microcontroller 10. The control unit 10 is designed to control the switching elements of the power electronics 3 in order to achieve a desired current supply to the phases of the stator winding.

[0029] The control unit 10 has a first computing unit 11 and a second computing unit 12. The first computing unit 11 has a current controller 13. The second computing unit 12 has a model predictive controller 14. The control unit 10 is designed to determine control signals for the switching elements of the power electronics 3 by means of the controllers 13 and 14 and to control the switching elements depending on the determined control signals, as described below with reference to Figure 2 is explained in more detail.

[0030] Figure 2 shows an advantageous method for operating the electrical machine 2 using a flow chart.

[0031] In a first step S1, the control unit 10 determines a target current vector i Soll,dq as a function of a predetermined target torque T Soll on the one hand and an actual rotation angle Θ Ist of the rotor on the other. The target torque T Soll is specified, for example, as a function of the actuation of an accelerator pedal of the motor vehicle. The actual rotation angle Θ Ist is detected, for example, by a rotation angle sensor assigned to the rotor. The target current vector i Soll,dq is a current vector related to a rotor-fixed coordinate system. The current vector describes the current value of a torque-generating current iq on the one hand and the current value of a flux-generating current id on the other. The current vector corresponds to an operating point of the electric machine. The target current vector i Soll,dq of the electric machine 2 is therefore the target operating point of the electric machine 2.

[0032] In a second step S2, the control unit 10 determines a difference between the target current vector i Soll,dq on the one hand and a determined actual current vector i Ist,dq on the other. The actual current vector i Ist,dq is determined, for example, as a function of the actual phase currents flowing through the phases of the motor winding using a d / q transformation. The actual current vector i Ist,dq corresponds to a current actual operating point of the electric machine 2.

[0033] In a third step S3, the control unit 10 uses the current controller 13 to determine a target voltage vector u Soll,dq related to the rotor-fixed coordinate system. The target voltage vector u Soll,dq describes terminal voltages that are to be applied to the phases of the stator winding in order to reduce the difference between the target current vector i Soll,dq and the actual current vector i Ist,dq.

[0034] In a fourth step S4, the control unit 10 determines control signals for the switching elements of the power electronics 3 as a function of the target voltage vector u Soll,dq.

[0035] In a fifth step S5, the control unit 10 controls the switching elements depending on the control signals determined in step S4.

[0036] The steps S1 to S5 are continuously carried out during normal operation of the electrical machine 2, so that a field-oriented control of the electrical machine 2 takes place by means of the steps S1 to S5.

[0037] In a sixth step S6, a threshold current value for the actual current vector i Ist,dq and a threshold voltage value for the terminal voltages are specified.

[0038] In a seventh step S7, the control unit 10 uses the model predictive controller 14 to determine a target trajectory for the actual current vector i Ist,dq , wherein the target trajectory runs from the current actual operating point of the machine 2 to a short-circuit operating point of the machine 2. The short-circuit operating point is the operating point of the machine 2 that the machine 2 assumes when the phases of the motor winding are short-circuited.

[0039] The model-predictive controller 14 determines the target trajectory in such a way that, by changing the actual current vector i Ist,dq along the target trajectory, a time-optimized setting of the short-circuit operating point is achieved starting from the current actual operating point. To this end, the model-predictive controller 14 determines the target trajectory as a function of a model of the electric machine 2.

[0040] The model-predictive controller 14 also takes the specified threshold current value into account when determining the target trajectory. To this end, the model-predictive controller 14 determines the target trajectory in such a way that the actual current vector i Ist,dq always falls below the threshold current value along the target trajectory when the short-circuit operating point is set, i.e., when the motor winding is short-circuited.

[0041] The model-predictive controller 14 also takes the specified threshold voltage value into account when determining the target trajectory. To this end, the model-predictive controller 14 determines the target trajectory in such a way that the terminal voltages always fall below the threshold voltage value when setting the short-circuit operating point along the target trajectory.

[0042] In an eighth step S8, the model-predictive controller 14 predicts a feedforward control depending on the determined target trajectory. In this case, the model-predictive controller 14 predicts a control sequence as feedforward control that has several optimized voltage vectors u opt,dq related to the rotor-fixed coordinate system. If the phases are sequentially applied with electrical terminal voltages according to the voltage vectors u opt,dq, the short-circuit operating point of the machine 2 is adjusted, starting from the current actual operating point of the machine 2, such that the profile of the actual current vector i Ist,dq at least substantially corresponds to the target trajectory.

[0043] Steps S6 to S8 are carried out continuously so that a target trajectory is always determined and a feedforward control is predicted for each current actual operating point of machine 2.

[0044] In a ninth step S9, the motor vehicle and the electric machine 2 are monitored for faults. If a fault is detected in step S9, the control signals are determined in step S4 as a function of the predicted feedforward control. Consideration of the target voltage vector u Soll,dq is suspended. Accordingly, in step S5, the switching elements are then controlled as a function of control signals that were determined as a function of the predicted feedforward control. This sets the short-circuit operating point in a time-optimized manner and without overloading the power electronics 3 and the motor winding.

[0045] Optionally, if a fault is detected in a tenth step S10, the energy storage device 4 is electrically disconnected from the power electronics 3. For this purpose, the battery contactor 7 and / or the battery contactor 8 are switched to non-conductive.

[0046] If the optional step S10 is carried out, a voltage threshold for an intermediate circuit voltage of the intermediate circuit 5 and a target voltage for the intermediate circuit voltage are preferably also specified in the sixth step S6.

[0047] The model-predictive controller 14 then also takes the specified voltage threshold into account when determining the target trajectory. To this end, the model-predictive controller 14 determines the target trajectory in such a way that the intermediate circuit voltage always falls below the voltage threshold when setting the short-circuit operating point along the target trajectory.

[0048] In addition, the model-predictive controller 14 also takes the specified target voltage into account when determining the target trajectory. To this end, the model-predictive controller 14 determines the target trajectory in such a way that the intermediate circuit voltage corresponds to the target voltage when the short-circuit operating point is reached.

[0049] According to a further exemplary embodiment, the desired trajectory is determined in preliminary tests, and the feedforward control is predicted in preliminary tests. In this case, the feedforward control is then stored in a data memory assigned to the machine 2. If an error is detected in this exemplary embodiment, the feedforward control is not predicted according to method steps S7 and S8, but is provided by the data memory. According to a further exemplary embodiment, the control signals for the switching elements are already determined in the preliminary tests and stored in the data memory instead of the feedforward control. Preferably, a desired trajectory is determined for each of several potential actual operating points, and a feedforward control is predicted, wherein the feedforward controls or the control signals are then preferably stored in a characteristic map.

[0050] Figure 3shows a current locus curve in which a first target trajectory T is represented. A voltage threshold for the intermediate circuit voltage and a target voltage for the intermediate circuit voltage were not taken into account when determining this first target trajectory T.

[0051] As from Figure 3 As can be seen, the threshold current value SSW is 400 amperes in this case. The current actual operating point AP1 corresponds to an intersection point of a first ISO torque characteristic curve KL1 with the MTPA curve. The current actual operating point AP1 is located in the third quadrant of the current locus. Machine 2 is therefore currently operating in generator mode. The short-circuit operating point AP2 lies on a second ISO torque characteristic curve KL2. The profile V of the actual current vector i Ist,dq essentially corresponds to the target trajectory T when setting the short-circuit operating point AP2 as a function of the feedforward control.

[0052] In Figure 4A left-hand diagram A and a right-hand diagram B are shown. In both diagrams A and B, the curve of the intermediate circuit voltage U ZK and the curve of a battery current i Bat are shown when setting the short-circuit operating point AP2 according to the first target trajectory T. In the case of the left-hand diagram A, the energy storage device 4 is electrically connected to the power electronics 3. In the case of the right-hand diagram B, the energy storage device 4 is electrically separated from the power electronics 3.

[0053] At a first time t1, the setting of the short-circuit operating point AP2 is initiated. The switching elements of the power electronics 3 are thus controlled from the first time t1 onwards depending on the predicted feedforward control. At a second time t2, the short-circuit operating point AP2 is reached. From this point onwards, the intermediate circuit voltage U ZK and the battery current i Bat are at least essentially constant. As can be seen from Figures A and B, only a time period of approximately 500 µs to 700 µs is required to set the short-circuit operating point AP2.

[0054] As can be seen from Figure A, the intermediate circuit voltage U ZK only increases by approximately 25 V during short-circuiting because, as the intermediate circuit voltage U ZK increases, charge carriers are supplied to the energy storage device 4 in the form of the battery current i Bat. If the energy storage device 4 is electrically connected to the power electronics 3, the short-circuit operating point AP2 can be set quickly and without overloading the intermediate circuit 5 according to the first target trajectory T.

[0055] As can be seen from Figure B, the intermediate circuit voltage U ZK increases significantly further when the energy storage device 4 is electrically separated from the intermediate circuit 5 than when the energy storage device 4 is electrically connected to the intermediate circuit 5.

[0056] Figure 5shows a current locus curve depicting a second target trajectory T. The voltage threshold for the intermediate circuit voltage U ZK and the target voltage for the intermediate circuit voltage U ZK were taken into account when determining the second target trajectory T by the model predictive controller 14. In this case, a voltage threshold of 480 V and a target voltage of 385 V were specified.

[0057] As from Figure 5 As can be seen, the second target trajectory T, starting from the actual operating point AP1, initially runs into the second quadrant of the current locus. This results in a reduction in the intermediate circuit voltage U ZK when the short-circuit operating point AP2 is set.

[0058] In Figure 6The curve of the intermediate circuit voltage U ZK and the curve of the battery current i Bat are shown when setting the short-circuit operating point AP2 according to the second target trajectory T. The energy storage device 4 is electrically separated from the power electronics 3. As can be seen from Figure 6 As can be seen, the intermediate circuit voltage U ZK nevertheless only temporarily rises to approximately 480 V and then drops to approximately 385 V. The intermediate circuit 5 is therefore not overloaded. The second target trajectory T is therefore particularly advantageous when the energy storage device 4 is electrically separated from the intermediate circuit 5.

[0059] Compared to the setting of the short-circuit operating point AP2 according to the first target trajectory T, a longer time is required for the setting of the short-circuit operating point AP2 according to the second target trajectory T. As can be seen from Figure 5As can be seen, the short-circuit operating point AP2 is reached at a third time t3. Setting the short-circuit operating point AP2 takes approximately 1.8 ms.

Claims

1. Method for operating an electric machine comprising a rotatably mounted rotor and a motor winding, wherein the motor winding is electrically connected to an electric energy store (4) by means of power electronics (3), wherein the machine (2) and / or a device comprising the machine (2) is / are monitored in respect of fault events, and wherein the motor winding is short-circuited by actuating the power electronics (3) upon detection of a fault event, characterized in that a target trajectory (T) for an actual current vector (iIst,dq) of an electric motor current flowing through the motor winding is ascertained, wherein the target trajectory (T) runs from a current actual operating point (AP1) of the machine (2) to a short-circuit operating point (AP2) of the machine (2), wherein a feedforward control action is predicted according to the target trajectory (T), and wherein the power electronics (3) are actuated to short-circuit the motor winding according to the feedforward control action in such a way that, when the motor winding is short-circuited, a curve (V) of the actual current vector (iIst,dq) at least substantially matches the target trajectory (T).

2. Method according to Claim 1, characterized in that the target trajectory (T) is ascertained according to the current actual operating point (AP1) during operation of the electric machine (2).

3. Method according to Claim 1, characterized in that the target trajectory (T) is ascertained in preliminary tests.

4. Method according to Claim 3, characterized in that a respective target trajectory (T) is ascertained for multiple potential actual operating points (AP1).

5. Method according to any one of the preceding claims, characterized in that the target trajectory (T) is ascertained according to a model of the electric machine (2).

6. Method according to any one of the preceding claims, characterized in that a threshold current value (SSW) is specified, wherein the target trajectory (T) is ascertained according to the threshold current value (SSW) in such a way that a current value of the actual current vector (iIst,dq) always undershoots the current threshold value (SSW) when the motor winding is short-circuited.

7. Method according to any one of the preceding claims, characterized in that a threshold voltage value is specified, wherein the target trajectory (T) is ascertained according to the threshold voltage value such that voltage values of electrical terminal voltages of the machine (2) always undershoot the threshold voltage value when the motor winding is short-circuited.

8. Method according to any one of the preceding claims, characterized in that the target trajectory (T) is ascertained according to a DC link voltage (UZK) of a DC link (5) of the power electronics (3).

9. Method according to Claim 8, characterized in that a voltage threshold for the DC link voltage (UZK) is specified, wherein the target trajectory (T) is ascertained according to the voltage threshold such that the DC link voltage (UZK) always undershoots the voltage threshold when the motor winding is short-circuited.

10. Method according to any one of Claims 8 and 9, characterized in that a target voltage for the DC link voltage (UZK) is specified, wherein the target trajectory (T) is ascertained according to the target voltage such that the DC link voltage (UZK) corresponds to the target voltage when the short-circuit operating point (AP2) is reached.

11. Method according to any one of the preceding claims, characterized in that the target trajectory (T) is ascertained by a model-predictive controller (14).

12. Method according to any one of the preceding claims, characterized in that a sensor signal of a sensor is compared with a predetermined limit value, and that the fault case is detected according to the comparison.

13. Method according to any one of the preceding claims, characterized in that the electric machine (2) is controlled at the short-circuit operating point according to the feedforward control action upon detection of the fault case.

14. Device for operating an electric machine, wherein the machine (2) has a rotatably mounted rotor and a motor winding, and wherein the motor winding is electrically connected to an electrical energy store (4) by power electronics (3), characterized by a control unit (10), in particular a microcontroller (10), which is specifically adapted to carry out the method according to any one of the preceding claims when used as intended.

15. Electric drive system, having an electric machine (2) comprising a rotatably mounted rotor and a motor winding, wherein the motor winding is electrically connected to an electrical energy store (4) by power electronics (3), characterized by a device (9) according to the preceding claim for operating the electric machine (2).