Method for operating a power converter, power converter for a permanently excited electric machine, vehicle and computer program product

Abstract

Method for operating a power converter for a permanently excited electric machine, wherein temperature information, which describes a temperature of at least one permanent magnet of the electric machine, is determined by means of an observer as a function of operating parameters of the electric machine, and the power converter is controlled as a function of the temperature information, wherein a computing device, which handles processes in time slices, carries out a first process in a first time slice for detecting parameter values for determining the operating parameters and carries out a second process, which determines the temperature information, in a second time slice, which is retrieved less frequently than the first time slice.

Claims

1. A method for operating a power converter for a permanently excited electric machine, wherein temperature information, which describes a temperature of at least one permanent magnet of the electric machine, is determined by means of an observer as a function of operating parameters of the electric machine, and the power converter is controlled as a function of the temperature information, wherein a computing device, which handles processes in time slices, carries out a first process in a first time slice for detecting parameter values for determining the operating parameters and carries out a second process, which determines the temperature information, in a second time slice, which is retrieved less frequently than the first time slice wherein the parameter values are detected over a duration of a detection cycle comprising one or more electrical or mechanical periods of the electric machine, and wherein the parameter values are only used to determine the operating parameters if a validation criterion describing the existence of quasi-stationary operation of the electric machine is met and a new detection cycle is carried out if the validation criterion is not met.

2. The method according to claim 1, wherein clocked switching signals for the control of the power converter are also generated by the first process.

3. The method according to claim 1, wherein mean values of the parameter values are calculated over the duration of the detection cycle for determining the operating parameters.

4. The method according to claim 1, wherein the validation criterion comprises a condition that a change of target currents of the power converter is within a specified range and/or a condition that a change of a speed of the electric machine is within a specified range.

5. The method according to claim 3, wherein the calculation of the mean values and/or an evaluation of the validation criterion is carried out by the second process or by a process carried out in a time slice which is retrieved less frequently than the first time slice.

6. The method according to claim 1, wherein the detection of the parameter values is started by trigger information which is only provided when start information is present which indicates that a speed of the electric machine is greater than or equal to a specified threshold value, and/or the determination of the operating parameters is started by trigger information which is provided only when the first process provides finish information, and/or the determination of the temperature information is started by trigger information which is provided only when the second process by which the operating parameters are calculated provides validation information.

7. The method according to claim 6, wherein the generation of the trigger information is controlled by the second process or by a process performed in a time slice which is retrieved less frequently than the first time slice.

8. The method according to claim 6, wherein the first process generating the trigger information realises a state machine which receives the start information and/or the finish information and/or the validation information as an input action.

9. The method according to claim 1, wherein, when determining the temperature information, a temperature difference to be added to a reference temperature is determined from a magnetic flux difference as a function of an operating parameter including actual output currents of the power converter.

10. A power converter for a permanently excited electric machine, comprising a computing device which is designed to carry out a method according to claim 1.

11. A vehicle comprising a permanently excited electric machine for driving the vehicle and a power converter according to claim 10, which is designed to supply the electric machine.

12. A computer program for loading into a memory of a computing device, comprising software code with which a method according to claim 1 is performed when the computer program is run on the computing device.

Description

(1) Further advantages and details of the present invention will become clear from the drawings described below. These are schematic representations and show:

(2) FIG. 1 a basic sketch of an embodiment of the vehicle according to the invention with an embodiment of the power converter according to the invention;

(3) FIG. 2 a flowchart of an embodiment of the method according to the invention;

(4) FIG. 3 a process diagram of processes carried out during the method; and

(5) FIG. 4 a block diagram of an observer used during the method.

(6) FIG. 1 is a schematic sketch of an embodiment of a vehicle 1, for example a hybrid vehicle or an electric vehicle, comprising a permanently excited electric machine 2 for driving the vehicle 1, an embodiment of a power converter 3 designed to supply the electric machine, and a high-voltage battery 4. The electric machine comprises a stator 5 and a rotor 7 having one or more permanent magnets 6.

(7) The power converter 3 has a power unit 8, which is designed to convert a DC voltage supplied by the high-voltage battery 4 into a multi-phase AC voltage, which is fed to the electric machine 2. The power unit 8 is controlled by a computing device 9, which is realised by a microcontroller, for example. The computing device 9 has a central computing unit 10, which is operated with a real-time operating system, and a memory 11. The computing device 9 handles processes of a computer program loaded into the memory 11 in a number of time slices of different retrieval frequency. The computing device 9 is thus designed to carry out a method for operating the power converter 3 according to one of the following embodiments:

(8) FIG. 2 is a block diagram of an embodiment of the method for operating the power converter 3. Within the scope of the method, a number of computing instructions or individual processes 12, 13, 14 are controlled by a higher-order individual process 15 and are presented in the process diagram shown in FIG. 3.

(9) The method is used to determine temperature information 16, which describes a mean value of the temperature distribution of the permanent magnets 6, and to control the power converter 3 as a function of the temperature information 16. This control is carried out in a further individual process 17. Only individual processes 12, 17 are carried out by a first process 18 in a first time slice, the retrieval frequency of which corresponds to the clock frequency of the power converter 3. At a clock frequency of 10 kHz, the first process 18 is retrieved every 100 μs, for example. The individual processes 13, 14, 15, however, are carried out by a second process 19 in a second time slice, which is retrieved less frequently than the first time slice, for example every 10 ms.

(10) In a step S1, the higher-level individual process 15 receives start information 20 as an input action, indicating that a speed of the electric machine 2 is greater than or equal to a specified threshold value, and then outputs trigger information 21. Only when this threshold value is reached or exceeded can the temperature information be meaningfully determined. For this purpose, the individual process 15 realises a state machine.

(11) In a subsequent step S2, the presence of trigger information 21 starts the individual process 12 for detection of parameter values 22, which is performed in the first time slice. The detection is performed over the duration of a detection cycle, which comprises one or more electrical or mechanical periods of the electric machine 2. During this detection cycle, the individual process 12 accumulates sample values as parameter values 22 for the determination of operating parameters that describe a speed of the electric machine 2, an actual output current of the power converter 3, a target current specified for controlling the power converter 3, and an output voltage of the power converter 3. In addition, the individual process 12 determines minimum and maximum values of the parameter values 22 for the determination of a particular operating parameter. The accumulated sample or parameter values 22 and also the minimum and maximum values are stored in the memory 11.

(12) In a subsequent step S3, the individual process 12 outputs finish information 23, which forms an input action of the state machine, to the individual process 15. If the finish information 23 is present, the individual process 15 outputs further trigger information 24.

(13) When trigger information 24 is present, the individual process 13 for determining the operating parameters, which is carried out in the second time slice, is started in a step S4. To determine the operating parameters, a mean value of the parameter values 22 is first calculated over the duration of the detection cycle by dividing the accumulated parameter values 22 retrieved from memory 11 by the number of samplings.

(14) In a subsequent step S5, the individual process 13 evaluates a validation criterion that describes a quasi-stationary operation of the electric machine 2. The validation criterion comprises a condition that a change in the target currents of the power converter 3 lies within a specified range, and a condition that a change in the speed of the electric machine lies within a specified range. For this purpose, the individual process 13 compares the minimum and maximum values, also retrieved from the memory 11, with specified threshold values.

(15) In a step S6, the individual process 13 checks whether the validation criterion is met. If this is not the case, negative validation information 25 is output as an input action for the state machine. The state machine then generates a new trigger information 21—delayed if necessary—which triggers a new detection cycle through the individual process 12. The method has a corresponding jump back to step S2.

(16) If, however, the check in the individual process 13 shows that the validation criterion is met, the mean values are thus stored as operating parameters in the memory 11 and positive validation information 26 is output as an input action to the superordinate individual process 15. This then generates further trigger information 27 for the individual process 14 in a step S7.

(17) If trigger information 27 is present, the individual process 14 for determining the temperature information 16, which is carried out in the second time slice, is started in a step S8. The process 14 realises an observer, which is shown in detail in FIG. 4.

(18) For this purpose, the individual process 14 comprises a sub-process 28, which retrieves the operating parameters, which comprise an averaged speed ω.sub.el,mean, an averaged q-component of the target or output voltage u.sub.q,mean and an averaged q-component of the actual output current i.sub.q,mean, from the memory 11. Based on the stationary voltage equation of the electric machine 2
u.sub.q=R.sub.q.Math.i.sub.q+ω.sub.el.Math.Ψ.sub.d
the sub-process 28 determines a d-component of the estimated magnetic flux

(19) ${\Psi }_{d,\mathrm{est}}=\frac{u_{q,\mathrm{mean}}-R_q.Math.i_{q,\mathrm{mean}}}{{\omega }_{\mathrm{el},\mathrm{mean}}}.$

(20) Here, R.sub.q describes a mean winding resistance effective in the q-axis, which is generally composed of an ohmic DC voltage component and an additional frequency-dependent component, which describes additional losses due to the skin effect and the proximity effect. The DC voltage component is adapted via a temperature coefficient of the conductor material and a measured winding temperature.

(21) A further sub-process 29 of the individual process 14 uses a lookup table to determine a d-component of a magnetic reference flux Ψ.sub.d,ref from averaged dq-components of the actual output current i.sub.dq,mean for a given reference temperature T.sub.PM,ref of the permanent magnets 6. The difference between the d-components of the estimated magnetic flux Ψ.sub.d,est and the magnetic reference flux Ψ.sub.d,ref results in a d-component of a magnetic flux difference ΔΨ.sub.d.

(22) A next sub-process 30 of the individual process 14 uses a lookup table to determine a temperature difference ΔT.sub.PM from the magnetic flux difference ΔΨ.sub.d and the averaged dq-components of the actual output current i.sub.dq,mean, and this temperature difference, when added to the reference temperature T.sub.PM,ref, gives the estimated temperature T.sub.PM,est of the permanent magnet 6 as temperature information 16.

(23) The determination of the estimated temperature T.sub.PM,est is based on the following Taylor series development:

(24) $T_{\mathrm{PM},\mathrm{est}}=T_{\mathrm{PM},\mathrm{ref}}+\frac{\partial T_{\mathrm{PM}}}{\partial {\Psi }_d}\left({\Psi }_{d,\mathrm{est}}-{\Psi }_{d,\mathrm{ref}}\right)+\frac{1}{2}\frac{{\partial }^2{T}_{\mathrm{PM}}}{\partial {\Psi }_d^2}{\left({\Psi }_{d,\mathrm{est}}-{\Psi }_{d,\mathrm{ref}}\right)}^2+\mathrm{.Math.}$

(25) The d-component of the magnetic reference flux Ψ.sub.d,ref thus describes the fundamental wave of the flux at the reference temperature T.sub.PM,ref. For the implementation described here, it is basically sufficient to consider only the first-order derivative for the estimation. The first-order derivative is basically a function of the dq components of the actual output current i.sub.dq, which is especially true for highly saturated traction machines in automotive applications. The first-order derivative can therefore be stored as a lookup table—as described above—or can be described by a polynomial depending on the averaged dq components of the actual output current i.sub.dq,mean.

(26) By taking into account the current averaged dq components of the actual output current i.sub.dq,mean, changes in the inductances that are ignored in conventional methods can also be taken into account. Especially in highly saturated machines, a change in the d-component of the magnetic flux dependent on the temperature of the permanent magnets 6 has a strong dependence on the dq-components of the actual output current, and thus their consideration leads to a significant increase in the accuracy of the estimation.

(27) In a last step S9, the power converter 3 is controlled by the individual process 17 carried out in the first time slice, depending on the temperature information 16. During this individual process, the clocked switching signals for switching elements of the power unit 8 are determined as a function of the temperature information 16, wherein a power reduction (derating) takes place if the temperature of the permanent magnets 6 is high. If this power reduction is not sufficient, protection of the permanent magnets 6 is initiated by reducing the speed of the electric machine 2. For this purpose a message is output to a superimposed control unit (not shown) of the vehicle 1. The control unit then implements a drivetrain limitation (speed limitation).

(28) According to further embodiments of the method, only one or some of the individual processes 13, 14, 15 are carried out in the first process 18, and the other individual processes are carried out as separate processes. Likewise, the individual processes 12, 17 can be carried out as separate processes. Here, it is merely essential that the time slices in which the processes corresponding to the individual processes 12, 17 are carried out are retrieved more frequently than the time slices in which the processes corresponding to the individual processes 13, 14, 15 are carried out.