METHOD FOR CONTROLLING A POWERTRAIN OF AN ELECTRIC VEHICLE, DATA PROCESSING DEVICE, COMPUTER PROGRAM, COMPUTER-READABLE MEDIUM, AND POWERTRAIN

Abstract

A method for controlling a powertrain of an electric vehicle. The method includes receiving a torque demand signal. Thereafter, a future power loss within the powertrain is estimated as a function of a torque distribution between at least two electric traction machines of the powertrain. Alternatively or additionally the loss can be estimated as a function of a free rolling state of at least one of the electric machines. Subsequently, a torque distribution between the electric traction machines is determined and/or a free rolling state of at least one of the electric machines is determined which minimizes the future power loss. Moreover, a corresponding data processing device, a corresponding computer program and a corresponding computer-readable medium are presented. Moreover, a powertrain for an electric vehicle is described. The powertrain includes such a data processing device and at least two electric traction machines and/or a clutch device.

Claims

1. A method for controlling a powertrain of an electric vehicle, the powertrain comprising at least two electric traction machines and/or a clutch device, the method comprising: receiving a torque demand signal being generated by a driver or a drive control unit; estimating a future power loss within the powertrain (10) as a function of a torque distribution between the at least two electric traction machines and/or as a function of a free rolling state of at least one of the at least two electric machines; and determining a torque distribution between the at least two electric traction machines and/or a free rolling state of at least one of the at least two electric machines which minimizes the future power loss.

2. The method according to claim 1, further comprising: deriving a torque demand signal for each of the at least two electric machines from the determined torque distribution and/or deriving a free rolling state signal for at least one of the at least two electric tractions machines and/or the clutch device; and providing the torque demand signals and/or the free rolling state signal to the corresponding electric machine and/or the clutch device.

3. The method according to claim 1, wherein estimating the future power loss comprises using a model describing at least partially the thermal behavior of the powertrain.

4. The method according to claim 3, wherein the model describes the thermal behavior of the powertrain by an electrical equivalent.

5. The method according to claim 4, wherein the thermal behavior of at least one component of the powertrain is described by an electric resistance as an equivalent of a thermal resistance and/or an electric capacitance as an equivalent of a thermal capacitance.

6. The method according to claim 1, wherein estimating the future power loss comprises calculating at least one of a future first inverter temperature (T.sub.inv1), a future first inverter entry temperature (Te.sub.2), a future second inverter temperature (T.sub.inv2), a future second inverter entry temperature (Te.sub.3), a future first electric traction machine temperature (T.sub.EM1), a future first electric traction machine entry temperature (Te.sub.4), a future second electric traction machine temperature (T.sub.EM2), a future second electric traction machine entry temperature (Te.sub.5), and a future radiator temperature (T.sub.radout).

7. The method according to claim 1, wherein estimating the future power loss comprises calculating a temporal differential of at least one of a first inverter temperature (T.sub.inv1), a first inverter entry temperature (Te.sub.2), a second inverter temperature (T.sub.inv2), a second inverter entry temperature (Te.sub.3), a first electric traction machine temperature (T.sub.EM1), a first electric traction machine entry temperature (Te.sub.4), a second electric traction machine temperature (T.sub.EM2), a second electric traction machine entry temperature (Te.sub.5), and a radiator temperature (T.sub.radout).

8. The method according to claim 1, further comprising: receiving at least one of an ambient temperature signal representing an ambient temperature (T.sub.amb), a battery voltage signal representing a current battery voltage (U.sub.DC), a vehicle speed signal representing a vehicle speed, a first rotational speed signal representing a rotational speed (ω.sub.1) of a first electric traction machine, a second rotational speed signal representing a rotational speed (ω.sub.2) of a second electric traction machine, a first torque signal representing a torque (M.sub.1) of a first electric traction machine, and a second torque signal representing a torque (M.sub.2) of a second electric traction machine.

9. The method according to claim 1, further comprising: receiving at least one of a first inverter temperature sensor signal, a first inverter entry temperature sensor signal, a second inverter temperature sensor signal, a second inverter entry temperature sensor signal, a first electric traction machine temperature sensor signal, a first electric traction machine entry temperature sensor signal, a second electric traction machine temperature sensor signal, a second electric traction machine entry temperature sensor signal, and a radiator temperature sensor signal.

10. The method according to claim 1, wherein the method is executed in real-time.

11. A data processing device comprising means comprising a processor executing instructions stored in a memory for carrying out the method according to claim 1.

12. A powertrain for an electric vehicle, comprising the data processing device according to claim 11, at least two electric traction machines and/or a clutch device, the data processing device being communicatively connected to each of the at least two electric traction machines and/or to the clutch device such that the at least two electric traction machines and/or the clutch device is controllable by the data processing device.

13. The powertrain according to claim 12, further comprising a battery system being electrically connected to the at least two electric traction machines, wherein the data processing device is embedded in the battery system.

14. A non-transitory computer-readable medium comprising instructions stored in a memory and executed by a processor to carry out steps for controlling a powertrain of an electric vehicle, the powertrain comprising at least two electric traction machines and/or a clutch device, the steps comprising: receiving a torque demand signal being generated by a driver or a drive control unit; estimating a future power loss within the powertrain (10) as a function of a torque distribution between the at least two electric traction machines and/or as a function of a free rolling state of at least one of the at least two electric machines; and determining a torque distribution between the at least two electric traction machines and/or a free rolling state of at least one of the at least two electric machines which minimizes the future power loss.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Examples of the disclosure will be described in the following with reference to the following drawings.

[0044] FIG. 1 shows an example powertrain according to the present disclosure, including a data processing device and a computer-readable medium according to the present disclosure, wherein the data processing device and the computer-readable medium are configured for running a computer program according to the present disclosure and for performing a method according to the present disclosure,

[0045] FIG. 2 shows a method for controlling a powertrain according to the present disclosure,

[0046] FIG. 3 shows a detail of a first example of the method of FIG. 2, and

[0047] FIG. 4 shows a detail of a second example of the method of FIG. 2.

[0048] The figures are merely schematic representations and serve only to illustrate examples of the disclosure. Identical or equivalent elements are in principle provided with the same reference signs.

DETAILED DESCRIPTION

[0049] FIG. 1 shows a powertrain 10 for an electric vehicle. The powertrain 10 includes two front wheels 12 and two rear wheels 14. Moreover, the powertrain 10 has a first electric traction machine 16 which is coupled to the front wheels 12 such that the first electric traction machine 16 can provide a propulsion torque to the front wheels. For controlling the first electric traction machine 16, a corresponding first machine control unit 16a is provided. The machine control unit 16a may be integrated into the electric traction machine 16. The powertrain 10 additionally includes a second electric traction machine 18. The second electric traction machine is coupled to the rear wheels 14 via a clutch device 20, a gearbox unit 22 and a differential gear 24. The clutch device 20 is controlled by a clutch device control unit 20a. Thus, the second electric traction machine 18 is configured for providing a propulsion torque to the rear wheels 14. For controlling the second electric traction machine 18, a corresponding second machine control unit 18a is provided. The second machine control unit 18a may be integrated into the electric traction machine 18.

[0050] The powertrain 10 also includes a battery system 26. The battery system 26 is electrically connected to the first electric traction machine 16 and the second electric traction machine 18 such that the first and second electric traction machines 16, 18 can be powered with energy being stored in the battery system 26.

[0051] Since in the present example, the first and second electric traction machines 16, 18 are AC (alternating current) machines, the first electric traction machine 16 is electrically connected to the battery system 26 via a first inverter unit 16b and the second electric traction machine 18 is electrically connected to the battery system 26 via a second inverter unit 18b.

[0052] It is noted that in alternative configurations, the inverter units 16b, 18b may be provided as parts of the corresponding electric machines 16, 18 or a corresponding control unit 16a, 18a. It is also possible that the electric machines 16, 18, the corresponding control units 16a, 18a and the corresponding inverter units 16b, 18b are integrated into one respective unit.

[0053] The battery system 26 is controlled by a battery system control unit 26a.

[0054] Furthermore, the powertrain 10 has a cooling system 28 which includes a cooling unit 30 having a coolant pump 32 and a radiator 34. Moreover, the cooling system 28 includes a coolant piping 36. Of course, the cooling system also includes a local control unit for controlling these components. The coolant piping 36 is configured such that the first electric traction machine 16, the first machine control unit 16a, the first inverter unit 16b, the second electric traction machine 18, the second machine control unit 18a, the second inverter unit 18b, and the battery system 26 can be provided with coolant in order to keep a temperature of these components within a desired range. The cooling unit 30 of the cooling system 28 is also electrically connected to the battery system 26 in order to be powered by the electric energy being stored in the battery system 26.

[0055] The powertrain 10 additionally has a powertrain control unit 38. This powertrain control unit 38 includes a data processing device 40 and a computer-readable medium 42.The data processing device 38 is able to control the first and second electric traction machines 16, 18 and the clutch device 20.

[0056] It is noted that in the present example, the clutch device 20 can be considered as a part of the second electric traction machine 18 even though it is represented as a separate part in FIG. 1.

[0057] The data processing device 40 and the computer readable medium 42 are configured for executing a computer program including instructions to carry out a method for controlling the powertrain 10. To this end, the powertrain control unit 38 is communicatively connected to the first machine control unit 16a, to the second machine control unit 18a, the battery control unit 26a and the cooling unit 30 of the cooling system 28. These connections are for example realized by a bus.

[0058] It is noted that the powertrain control unit 38 is represented outside the battery system 26 for better visibility. It can as well be integrated into the battery system 26. More generally speaking, the powertrain control unit 38 may be integrated into any another control unit of the powertrain 10.

[0059] The powertrain 10 being represented in FIG. 1 is to be seen as an example only. Of course, variations thereof are also included by the present disclosure. In a non-represented alternative, each of the wheels 12, 14 may be equipped with a dedicated electric traction machine and a dedicated clutch device. In other words, in an alternative powertrain, one electric traction machine and one clutch device is provided per wheel. In a further example, a total of three electric traction machines are provided, one being coupled to the front wheels, i.e. the front axle, and the remaining two being coupled to one rear wheel respectively. Also the opposite is possible, i.e. one electric traction machine being coupled to the rear wheels, i.e. the rear axle, and the remaining two being coupled to one front wheel respectively.

[0060] In the following, the method for controlling the powertrain 10 will be explained with additional reference to FIG. 2.

[0061] In a first step S1, a torque demand signal is received by the powertrain control unit 38. In the present example, this signal is generated by a driver pressing a drive pedal.

[0062] Subsequently, in a second step S2, a future power loss within the powertrain 10 is estimated as a function of a torque distribution between the first electric traction machine 16 and the second electric traction machine 18 and as a function of a free rolling state of the second electric traction machine 18 which is realized by the clutch device 20. The torque distribution defines how much torque is requested from the first electric traction machine 16 and how much torque is requested from the second electric traction machine 18. In other words, the torque distribution describes how the torque to be provided is split between the first electric traction machine 16 and the second electric traction machine 18. The torque distribution may, thus, also be called a torque split.

[0063] In the present example, the second electric traction machine 18 is in a free rolling state if the clutch device 20 is in an open, disengaged condition. Thus, in other words the future power loss is estimated as a function of an opening state of the clutch device 20.

[0064] In the present example, a model is used for determining the future power loss. This model describes the thermal behavior of the powertrain 10. A first example of this model is illustrated in FIG. 3 wherein the model describes the thermal behavior of the powertrain 10 by an electrical equivalent. In more detail: The thermal model shown in FIG. 3 includes four sub-models (a), (b), (c), and (d) being represented as electric circuits. The sub-model (a) relates to the first inverter unit 16b. In this context, a thermal resistance Rinvi of the first inverter unit 16b is described by an electric resistance. A thermal capacitance Cinvi of the first inverter unit 16b is described using an electric capacitance. An ambient temperature T.sub.amb is seen as an equivalent to an electric voltage. The same applies to a first inverter temperature T.sub.inv1 of the first inverter unit 16b. A power loss P.sub.loss, inv1 is described as an electric current. Due to the fact that this model is an electric representation, Kirchhoff’s law can be applied according to which

[00001]$i_1+i_2+i_3=0$

[0065] Consequently, the following differential equation can be used for describing the thermal behavior of the first inverter unit 16b. in this equation d/dt denotes the temporal differential:

[00002]$P_{loss,inv1}-C_{inv1}\frac{d{T}_{inv1}}{dt}+\frac{T_{amb}-T_{inv1}}{R_{inv1}}=0$

[00003]$C_{inv1}\frac{d{T}_{inv1}}{dt}=P_{loss,inv1}+\frac{T_{amb}-T_{inv1}}{R_{inv1}}$

[0066] The sub-model (b) relates to the second inverter unit 18b which is modelled in the same manner as the first inverter unit 16b. Thus, the thermal behavior of the second inverter unit 18b can be described by the following equation:

[00004]$C_{inv2}\frac{d{T}_{inv2}}{dt}=P_{loss,inv2}+\frac{T_{amb}-T_{inv2}}{R_{inv2}}$

[0067] The sub-model (c) relates to the first electric traction machine 16. Its thermal behavior is modelled in the same manner as for the inverter units 16b, 18b. Thus, the thermal behavior of the first electric traction machine 16 can be described by the following equation:

[00005]$C_{EM1}\frac{d{T}_{EM1}}{dt}=P_{loss,\text{EM}1}+\frac{T_{amb}-T_{EM1}}{R_{EM1}}$

[0068] The sub-model (d) relates to the second electric traction machine 18 which is modelled in the same manner as the inverter units 16b, 18b and the first electric traction machine 16. Thus, the thermal behavior of the second electric traction machine 18 can be described by the following equation:

[00006]$C_{EM2}\frac{d{T}_{EM2}}{dt}=P_{loss,\text{EM}2}+\frac{T_{amb}-T_{EM2}}{R_{EM2}}$

[0069] The above equations can be summarized in the following equation:

[00007]$\dot{T}=AT+BU$

[0070] In this equation

[00008]$T=\left[\begin{array}{c}{T}_{EM1}\\ T_{EM2}\\ T_{inv1}\\ T_{inv2}\end{array}\right]$

[0071] A is the following matrix:

[00009]$A=\left[\begin{array}{cccc}-\frac{1}{R_{EM1}{C}_{EM1}}& 0& 0& 0\\ 0& -\frac{1}{R_{EM2}{C}_{EM2}}& 0& 0\\ 0& 0& -\frac{1}{R_{inv1}{C}_{inv1}}& 0\\ 0& 0& 0& -\frac{1}{R_{inv2}{C}_{inv2}}\end{array}\right]$

[0072] B is the following matrix:

[00010]$B=\left[\begin{array}{ccccc}\frac{1}{C_{EM1}}& 0& 0& 0& \frac{1}{R_{EM1}{C}_{EM1}}\\ 0& \frac{1}{C_{EM2}}& 0& 0& \frac{1}{R_{EM2}{C}_{EM2}}\\ 0& 0& \frac{1}{C_{inv1}}& 0& \frac{1}{R_{inv1}{C}_{inv1}}\\ 0& 0& 0& \frac{1}{C_{inv2}}& \frac{1}{R_{inv2}{C}_{inv2}}\end{array}\right]$

[0073] U is the following vector:

[00011]$U=\left[\begin{array}{c}{P}_{loss,EM1}\\ P_{loss,EM2}\\ P_{loss,inv1}\\ P_{loss,inv2}\\ T_{amb}\end{array}\right]$

[0074] In this model, the temperatures T.sub.inv1, T.sub.inv2, T.sub.EM1, and T.sub.EM2 are considered as variables. The remaining parameters are fixed and known. Consequently, the model according to FIG. 3 can be designated a four-state model.

[0075] Alternatively, a model as shown in FIG. 4 can be used. Also this model describes the thermal behavior of the powertrain 10 by an electrical equivalent. In contrast to the model of FIG. 3, now the electric circuits by which the first electric traction machine 16, the second electric traction machine 18, the first inverter unit 16b and the second inverter unit 18b are described are connected. In more detail, the first electric traction machine 16 and the corresponding first inverter unit 16b are connected in series. Also the second electric traction machine 18 and the corresponding second inverter unit 18b are connected in series. The partial circuits including one of the electric traction machines 16, 18 and the corresponding inverter unit 16b, 18b are connected in parallel. The resulting circuit is connected to a partial circuit describing the radiator 34.

[0076] It is noted that the above-described structure is of course due to the architecture of the specific coolant circuit in the present example. For coolant circuits having a different layout, also a different corresponding electric structure would be used.

[0077] In this model, a heat flow is identified with an electric current. In FIG. 4 the heat flow is indicated by two arrows 44.

[0078] In this example, the coolant circuit being represented by the electrical equivalent of FIG. 4 is divided into five temperature zones. Thus, in addition to the temperatures T.sub.inv1, T.sub.inv2, T.sub.EM1, and T.sub.EM2, the model according to FIG. 4 considers a first inverter outlet temperature Te.sub.2, a second inverter outlet temperature Te.sub.3, a first electric traction machine outlet temperature Te.sub.4, and a second electric traction machine outlet temperature Tes as variables. These outlet temperatures refer to a temperature of a coolant at an outlet of the respective component of the powertrain 10. Also a radiator outlet temperature T.sub.radout is considered. This temperature describes a temperature of a coolant when leaving the radiator 34.

[0079] Thus, in the model of FIG. 4, nine temperatures are considered to be variables. The model can be designated a nine-state model. As before, the remaining parameters are considered to be fixed and known. In the model of FIG. 4 it is further assumed that at a connection point where the coolant coming from the first electric traction machine 16 and the second electric traction machine 18 mixes, the resulting temperature corresponds to an average of the two temperatures Te.sub.4 and Tes.

[0080] When applying the same modeling principles as in the model of FIG. 3, cf. Kirchhoff’s law above, and the first law of Thermodynamics, the thermal behavior of the powertrain 10 can be described by the following differential equations, wherein m is the mass flow of coolant, C.sub.p is the thermal capacity of the coolant, and h is the mass specific enthalpy. In the representation of FIG. 4, Kirchhoff’s law can be applied to every point of the circuit where the reference lines of T.sub.radout, Te.sub.2, Te.sub.3, Te.sub.4 and Tes end, i.e. in each of these points the sum of currents is zero. For the point where the reference line of Te.sub.2 ends, the following equations result:

[00012]$i_1+i_2+i_3=0$

[00013]$-C_2\cdot \frac{dT{e}_2}{dt}+\frac{T_{inv}-T{e}_2}{R_{inv}}+\dot{m}\cdot C_p\cdot \left(T_{radout}-T{e}_2\right)=0$

[00014]$C_2\cdot \frac{dT{e}_2}{dt}=-\dot{m}\cdot C_p\cdot \left(T{e}_2-T_{radout}\right)+\frac{T_{inv}-T{e}_2}{R_{inv}}$

[0081] When proceeding in the same manner for the other points, the following differential equations can be given:

[00015]$\begin{array}{c}{C}_{radout}\frac{d{T}_{radout}}{dt}=\dot{m}{c}_p\left(\frac{T{e}_4+T{e}_5}{2}-T_{radout}\right)-h\left(T_{radout}-T_{amb}\right)\\ C_2\frac{dT{e}_2}{dt}=-\frac{\dot{m}}{2}{c}_p\left(T{e}_2-T_{radout}\right)+\frac{T_{inv}-T{e}_2}{R_{inv1}}\\ C_3\frac{dT{e}_3}{dt}=-\frac{\dot{m}}{2}{c}_p\left(T{e}_3-T_{radout}\right)+\frac{T_{inv}-T{e}_3}{R_{inv2}}\\ C_4\frac{dT{e}_4}{dt}=-\frac{\dot{m}}{2}{c}_p\left(T{e}_4-T{e}_2\right)+\frac{T_{EM1}-T{e}_4}{R_{EM1}}\\ C_5\frac{dT{e}_5}{dt}=-\frac{\dot{m}}{2}{c}_p\left(T{e}_5-T{e}_3\right)+\frac{T_{EM2}-T{e}_5}{R_{EM2}}\\ C_{EM1}\frac{\text{d}{T}_{EM1}}{\text{d}t}=\frac{T_{EM1}-T{e}_4}{R_{EM1}}+P_{loss,EM1}\\ C_{EM2}\frac{\text{d}{T}_{EM2}}{\text{d}t}=\frac{T_{EM2}-T{e}_5}{R_{EM2}}+P_{loss,EM2}\\ C_{inv1}\frac{\text{d}{T}_{inv1}}{\text{d}t}=\frac{T_{inv1}-T{e}_2}{R_{inv1}}+P_{loss,inv1}\\ C_{inv2}\frac{\text{d}{T}_{inv2}}{\text{d}t}=\frac{T_{inv2}-T{e}_3}{R_{inv2}}+P_{loss,inv2}\end{array}$

[0082] These equations can be summarized as follows.

[00016]$\dot{T}=C^{-1}\left[AT+BU\right]$

[0083] In this equation, T is again the following vector:

[00017]$T=\left[\begin{array}{ccccccccc}{T}_{radout}& T{e}_2& T{e}_3& T{e}_4& T{e}_5& T_{EM1}& T_{EM2}& T_{inv1}& T_{inv2}\end{array}\right].$

[0084] A, B and C are matrices reading:

[00018]$A=\left[\begin{array}{ccccccccc}-\dot{m}{c}_p-h& 0& 0& \frac{\dot{m}{c}_p}{2}& \frac{\dot{m}{c}_p}{2}& 0& 0& 0& 0\\ \frac{\dot{m}{c}_p}{2}& \left(\frac{-\dot{m}{c}_p}{2}-\frac{1}{R_{inv1}}\right)& 0& 0& 0& 0& 0& \frac{1}{R_{inv1}}& 0\\ \frac{\dot{m}{c}_p}{2}& 0& \left(\frac{-\dot{m}{c}_p}{2}-\frac{1}{R_{inv2}}\right)& 0& 0& 0& 0& 0& \frac{1}{R_{inv2}}\\ 0& \frac{\dot{m}{c}_p}{2}& 0& \left(\frac{-\dot{m}{c}_p}{2}-\frac{1}{R_{EM1}}\right)& 0& \frac{1}{R_{EM1}}& 0& 0& 0\\ 0& 0& \frac{\dot{m}{c}_p}{2}& 0& \left(\frac{-\dot{m}{c}_p}{2}-\frac{1}{R_{EM2}}\right)& 0& \frac{1}{R_{EM2}}& 0& 0\\ 0& 0& 0& \frac{1}{R_{EM1}}& 0& \frac{-1}{R_{EM1}}& 0& 0& 0\\ 0& 0& 0& 0& \frac{1}{R_{EM2}}& 0& \frac{-1}{R_{EM2}}& 0& 0\\ 0& \frac{1}{R_{inv1}}& 0& 0& 0& 0& 0& \frac{-1}{R_{inv1}}& 0\\ 0& 0& \frac{1}{R_{inv2}}& 0& 0& 0& 0& 0& \frac{-1}{R_{inv2}}\end{array}\right]$

[00019]$B=\left[\begin{array}{ccccc}0& 0& 0& 0& h\\ 0& 0& 0& 0& 0\\ 0& 0& 0& 0& 0\\ 0& 0& 0& 0& 0\\ 0& 0& 0& 0& 0\\ 1& 0& 0& 0& 0\\ 0& 1& 0& 0& 0\\ 0& 0& 1& 0& 0\\ 0& 0& 0& 1& 0\end{array}\right]$

[00020]$C=\left[\begin{array}{ccccccccc}{C}_{radout}& 0& 0& 0& 0& 0& 0& 0& 0\\ 0& C_2& 0& 0& 0& 0& 0& 0& 0\\ 0& 0& C_3& 0& 0& 0& 0& 0& 0\\ 0& 0& 0& C_4& 0& 0& 0& 0& 0\\ 0& 0& 0& 0& C_5& 0& 0& 0& 0\\ 0& 0& 0& 0& 0& C_{EM1}& 0& 0& 0\\ 0& 0& 0& 0& 0& 0& C_{EM2}& 0& 0\\ 0& 0& 0& 0& 0& 0& 0& C_{inv1}& 0\\ 0& 0& 0& 0& 0& 0& 0& 0& C_{inv2}\end{array}\right]$

[0085] As before, U is a vector:

[00021]$U=\left[\begin{array}{c}{P}_{loss,EM1}\\ P_{loss,EM2}\\ P_{loss,inv1}\\ P_{loss,inv2}\\ T_{amb}\end{array}\right]$

[0086] A current power loss within the powertrain 10 can be estimated by summing up the losses of the components forming the powertrain 10, i.e. the loss of the first electric traction machine 16, the loss of the second electric traction machine 18, the loss of the first inverter unit 16b and the loss of the second inverter unit 18b (cf. formula below). These losses are each a function of the temperature of the respective component, a torque being provided by the respective component, a rotational speed being provided by the respective component and a voltage of the battery system 26.

[0087] Thus, the current power loss of the powertrain 10 can be represented by the following formula, wherein k is a time instant. The below equation represents an objective function in which the losses for all the time steps in the horizon k= 0 to k=N are cumulated:

[00022]$\underset{\lambda ,c}{Minimize}{\displaystyle \underset{k=0}{\overset{N}{.Math.}}\left(\begin{array}{c}{P}_{loss,EM1,k}\left(M_{1,k},{\omega }_{1,k},T_k,U_{DC,k}\right)+P_{loss,Em,2,k}\left(M_{2,k},{\omega }_{2,k},T_k,U_{DC,k}\right)+.Math.\\ +P_{loss,Inv1,k}\left(M_{1,k},{\omega }_{1,k},T_k,U_{DC,k}\right)+P_{loss,Inv,2,k}\left(M_{2,k},{\omega }_{2,k},T_k,U_{DC,k}\right)\end{array}\right)}$

[00023]$T_k=\left[\begin{array}{c}{T}_{rad\_out}\\ T_{EM1}\\ T_{EM1}\\ T_{inv1}\\ T_{inv2}\end{array}\right].$

[0088] The following conditions apply

[00024]$\left(\begin{array}{l}{M}_{1,k}={\lambda }_k{M}_{demand,k}\hfill \\ M_{2,k}=\left(1-{\lambda }_k\right)M_{demand,k}\hfill \\ {\omega }_{2,k}=c_k{v}_k/r_{wheel}\hfill \\ \max \left(-M_{EM1,\mathrm{l}\text{imit}},-M_{EM1,\text{limit}}\left({\omega }_{1,k}\right)\right)\le M_{1,k}\le \min \left(M_{EM1,\text{limit}},M_{EM1,}{}_{\text{limit}}\left({\omega }_{1,k}\right)\right)\hfill \\ \max \left(-M_{EM2,\mathrm{l}\text{imit}},-M_{EM2,\text{limit}}\left({\omega }_{2,k}\right)\right)\le M_{2,k}\le \min \left(M_{EM2,\text{limit}},M_{EM2,}{}_{\text{limit}}\left({\omega }_{2,k}\right)\right)\hfill \\ {\lambda }_k\in \left[0,1\right]\text{Torque}\text{split}\hfill \\ c_k\in \left\{0,1\right\}\text{Clutch},0.Math.\text{disengage},1.Math.\text{engage}\hfill \end{array}\right\}k=0,\mathrm{...},N-1$

[0089] In this context, λ.sub.k is the torque distribution or the torque split. The variable c.sub.k describes the opening state of the clutch device 20. It is noted again that for the present explanations k is zero, i.e. the current moment.

[0090] Concerning the calculation of the above power losses P.sub.loss, .sub.EM1,k, P.sub.loss,EM2,k, P.sub.loss,inv1,k, and P.sub.loss,inv2,k, it has been found that the power loss of the electric traction machine 16 and 18 is a function of torque, motor speed, DC Voltage, temperature of rotor and end-winding. The power loss of the inverter units 16b, 18b is a function of torque, motor speed, DC Voltage, temperature of rotor, end-winding and inverter switching unit.

[0091] The power losses of the electric traction machines 16, 18 and the power losses of the inverter units 16b, 18b can be implemented as 5-D/6-D lookup tables. These look-up tables may be created by performing experiments in a laboratory or by performing computer simulations. Of course, also a combination of experiments and simulations is possible.

[0092] In order to further reduce computational resources, the 5D/6D look-up tables can be simplified to three dimensions and a polynomial curve-fit equation can be created.

[0093] For calculating the current power loss, at least one of the following parameters may be detected by a sensor of the powertrain 10 and provided to the powertrain control unit 38: an ambient temperature signal representing an ambient temperature T.sub.amb, a battery voltage signal representing a current battery voltage U.sub.DC, a first rotational speed signal representing a rotational speed ω.sub.1 of the first electric traction machine 16, a second rotational speed signal representing a rotational speed ω.sub.2 of the second electric traction machine 18, a first torque signal representing a torque M.sub.1 of the first electric traction machine 16, and a second torque signal representing a torque M.sub.2 of a second electric traction machine 18.

[0094] It is noted that the rotational speeds ω.sub.1, ω.sub.2 can be calculated if a vehicle speed v.sub.k and a radius of the corresponding wheel r.sub.wheel is known.

[0095] For the current moment, described by k=0, also at least one of the following sensor signals may be received by the powertrain control unit 38: a first inverter temperature sensor signal representing the first inverter temperature Tinvi, a first inverter entry temperature sensor signal representing the first inverter entry temperature Te.sub.2, a second inverter temperature sensor signal representing the second inverter temperature T.sub.inv2, a second inverter entry temperature sensor signal representing the second inverter entry temperature Te.sub.3, a first electric traction machine temperature sensor signal representing the first electric traction machine temperature T.sub.EM1, a first electric traction machine entry temperature sensor signal representing the first electric traction machine entry temperature Te.sub.4, a second electric traction machine temperature sensor signal representing the second electric traction machine temperature T.sub.EM2, a second electric traction machine entry temperature sensor signal representing the second electric traction machine entry temperature Tes, and a radiator temperature sensor signal representing the radiator temperature Tradout.

[0096] It is noted that all of the above sensor signals may be used if the model according to FIG. 4 is used. If the model according to FIG. 3 is used, just the relevant sensor signals are used, i.e. just the sensor signals having an equivalent in the model of FIG. 3.

[0097] Based thereon and using one of the above-described models a temporal differential d/dt of at least one of the first inverter temperature T.sub.inv1, the first inverter entry temperatureTe.sub.2, a second inverter temperature T.sub.inv2, a second inverter entry temperature Te.sub.3, a first electric traction machine temperature T.sub.EM1, a first electric traction machine entry temperature Te.sub.4, a second electric traction machine temperature T.sub.EM2, a second electric traction machine entry temperature Te.sub.5, and a radiator temperature T.sub.radout can be calculated. It is noted that in a practical implementation, the differential may be discretized.

[0098] It is noted again that all of the above differentials may be calculated if the model according to FIG. 4 is used. If the model according to FIG. 3 is used, just the relevant differentials are calculated.

[0099] Using the temporal differentials, at least one of a future first inverter temperature T.sub.inv1, a future first inverter entry temperature Te.sub.2, a future second inverter temperature T.sub.inv2, a future second inverter entry temperature Te.sub.3, a future first electric traction machine temperature T.sub.EM1, a future first electric traction machine entry temperature Te.sub.4, a future second electric traction machine temperature T.sub.EM2, a future second electric traction machine entry temperature Tes, and a future radiator temperature T.sub.radout can be calculated.

[0100] Again, the only the future temperatures being relevant for the selected model are calculated. The future temperatures are calculated according to the following formula, wherein the matrices A and B of the relevant model are used:

[00025]$T_{k+1}=A{T}_k+B{u}_{k,}$

[00026]$k=0,\mathrm{...},N-1$

[0101] Thus, starting from the current moment at k=0, future temperatures can be calculated in an iterative manner, wherein k defines a step into the future. This means that starting from k=0, the temperatures at k=1 are calculated. Then, the temperatures at k=2 are calculated using the temperatures at k=1, and so on.

[0102] The prediction horizon N may be specifically chosen for each application scenario. It is noted that increasing N increases the computational performance requirements for performing the method.

[0103] Consequently, in a third step S3, a torque distribution λ.sub.k between the electric traction machines 16, 18 and a closing state c.sub.k of the clutch device 20 can be determined which minimizes the future power loss, i.e. the power loss at k>0, more precisely at k=N-1.

[0104] As has already been explained before, the closing state c.sub.k of the clutch device 20 describes a free rolling state of the second electric traction machine 18.

[0105] Once this is known, in a fourth step S4, torque demand signal for each electric machine 16, 18 may be derived and provided to the corresponding electric machine 16, 18.

[0106] In the same manner, a free rolling state signal can be derived and provided to the clutch device 20.

[0107] The method according to the present disclosure may be executed in real-time.

[0108] In the present example, the torque distribution λ.sub.k and the closing state c.sub.k may be updated with a predefined temporal frequency. This leads to a temporal interval between the corresponding torque demand signals and the free rolling state signal.

[0109] The real-time constraint in this context may be that the execution of the method needs to be terminated within 25% of the temporal interval. Consequently, one can be sure that the torque demand signals and the free rolling state signal are always chosen such that the powertrain 10 is operated at enhanced energy efficiency.

[0110] In connection with FIGS. 3 and 4 a four-state model and a nine-state model have been described. These models are used for performing the method according to the present disclosure. Of course also models having a different number of states may be used. Such models may be derived from the above four-state model and nine-state model by applying model order reduction techniques thereto. Models of reduced order may be implemented and executed with reduced computational complexity. However, it is important to note that states resulting from such a model order reduction technique may be fictitious states, i.e. they do not have any physical meaning.

[0111] Other variations to the disclosed examples can be understood and effected by those skilled in the art in practicing the claimed disclosure, from the study of the drawings, the disclosure, and the appended claims. In the claims the word “comprising” does not exclude other elements or steps and the indefinite article “a” or “an” does not exclude a plurality. A single processor or other unit may fulfill the functions of several items or steps recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored/distributed on a suitable medium such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope of the claims.