DE102015108067A1 - Method for starting an internal combustion engine - Google Patents

Abstract

The invention relates to a method for starting an internal combustion engine (1) of a hybrid vehicle during at least one purely electric drive mode, wherein in at least one phase of starting a between the internal combustion engine (1) and an electric machine (3) arranged separating clutch (2) at least partially closed and the internal combustion engine (1) by the electric machine (3) is entrained. In order to enable a smooth start of the internal combustion engine (1) in a hybrid vehicle, in particular while driving, it is provided that the torque fluctuations occurring on at least one drive wheel (7) during towing of the internal combustion engine (1) by means of a drive train model (9 ) - preferably continuously - calculated and predicted for the entire Mitschleppvorgang the internal combustion engine (1), and that the torque fluctuations occurring on the basis of the predicted torque fluctuations actively by at least one opposing correction torque (MKorr) at least reduced, preferably eliminated.

Description

The invention relates to a method for starting an internal combustion engine of a hybrid vehicle during at least one purely electric drive mode, wherein in at least one phase of starting a arranged between the internal combustion engine and an electric machine clutch is at least partially closed and the internal combustion engine is dragged by the electric machine.

By using the hybrid functions (stop / start, recuperation, load point increase or the like) an energy-efficient operation can be achieved. In hybrid vehicles, which have an internal combustion engine and at least one electric machine as drive machines, the internal combustion engine is frequently stopped at standstill and during a purely electric drive.

• • Eine zweite im Antriebsstrang verbaute Kupplung in Schlupf bringen, wie zum Beispiel in der DE 10 2006 034 937 A1 beschrieben. Dabei wird vor dem Starten des Verbrennungsmotors das vom Elektromotor in den Antriebsbtrang geleitete Drehmoment derart erhöht, dass in den Getriebeeingang nur das aktuelle Wunschmoment eingeleitet wird. Durch Ansteuern einer den Verbrennungsmotor mit dem Elektromotor verbindenden ersten Kupplung wird dem Antriebsstrang überschüssiges Drehmoment entzogen und in den Verbrennungsmotor zu dessen Beschleunigung eingeleitet. Bei Erreichen seiner Startdrehzahl wird der Verbrennungsmotor gezündet. Zur Reduzierung der Drehmomentenübertragung auf den Getriebeeingang wird eine den Elektromotor mit dem Getriebe verbindende zweite Kupplung derart angesteuert, dass in den Getriebeeingang nur das aktuelle Wunschmoment eingeleitet wird.
• • Änderung des Übersetzungsverhältnisses, wie zum Beispiel in der DE 10 2011 002 742 A1 erläutert. Dabei wird zum Starten des Verbennungsmotors die zwischen Verbrennungsmotor und elektrischer Maschine geschaltete Kupplung zumindest teilweise geschlossen und der Verbrennungsmotor übetr die elektrische Maschine angeschleppt. Parallel zum Starten des Verbennungsmotors wird mit der Ausführung einer Rückschaltung im Schaltgetriebe dann begonnen, wenn eine Drehzahl des Verbennungsmotors einen applizierbaren Grenzwert erreicht oder überschreitet.

In order to ensure a constant torque on the wheel during the hybrid start, that is to say starting the internal combustion engine by means of the electric drive machine, of a parallel hybrid drive having an internal combustion engine and at least one electric machine, two solutions are known:

• • Slip a second clutch installed in the drivetrain, such as the DE 10 2006 034 937 A1 described. In this case, prior to starting the internal combustion engine, the torque conducted by the electric motor into the drive train is increased such that only the current desired torque is introduced into the transmission input. By driving a first clutch connecting the internal combustion engine with the electric motor, excess torque is withdrawn from the drive train and introduced into the internal combustion engine for its acceleration. Upon reaching its starting speed of the engine is ignited. To reduce the transmission of torque to the transmission input, a second clutch connecting the electric motor to the transmission is activated such that only the current desired torque is introduced into the transmission input.
• • change of gear ratio, such as in the DE 10 2011 002 742 A1 explained. In this case, to start the Verbennungsmotors switched between the engine and electric machine clutch is at least partially closed and the engine übetr the electric machine towed. Parallel to starting the Verbennungsmotors is started with the execution of a downshift in the manual when a speed of the Verbennungsmotors reaches or exceeds an applicable limit.

For both of these solutions, an automatic transmission, for example, CVT, dual-clutch transmission, or the like, is needed. For a manually operated transmission these solutions are not usable because the second clutch and the gear ratio are manipulated by the driver.

From the DE 10 2011 109 353 A1 is a method for operating a railless land vehicle with an internal combustion engine and designed for selecting different ratios gear known, which is coupled on the input side by means of at least one clutch to an output shaft of the internal combustion engine and the output side connected to the drive wheels of the land vehicle. In a sailing operation of the land vehicle in which the land vehicle rolls with the combustion engine switched off and the clutch disengaged, the internal combustion engine is towed by closing the clutch. The towing process of the internal combustion engine includes the following steps: In a first step, the clutch is controlled from its open state with a first torque gradient at least as far closed until its clutch torque exceeds the drag torque of the internal combustion engine. In a second method step, the clutch is closed with a second torque gradient at least until the speed of the internal combustion engine exceeds a resonance speed of a two-mass flywheel arranged between the clutch and the output shaft of the internal combustion engine. In a third step, the clutch is fully closed in a controlled manner so that the speed of the internal combustion engine of the input-side speed of the transmission is smoothly adjusted.

The DE 198 14 402 A1 describes a drive system for a motor vehicle with an internal combustion engine and at least one electric machine, wherein the start-up phase of the vehicle runs so that the vehicle is initially accelerated solely by the electric machine, the internal combustion engine is started during which and then takes over the drive of the vehicle. A jerky coupling of the internal combustion engine is to be avoided by the fact that the internal combustion engine, while the electric machine accelerates the vehicle is entrained, or the engine is rotated in decoupled from the drive state for the purpose of starting and coupled at synchronous speed to the drive. Torque variations occurring during towing the internal combustion engine are actively reduced by opposing torques applied by an electric machine become. In particular, the opposing torques are applied by the electric machine driving the vehicle and thereby superimposed on the driving moment.

Furthermore is from the DE 10 2006 047 655 A1 a method for operating a parallel hybrid drive of a vehicle with an electric machine and a Verbennungsmotor known, wherein in the driving state of the vehicle, a start of the internal combustion engine is performed by means of the electric machine by closing a separating clutch. In this case, at least one operating variable of the parallel hybrid drive is detected and compared with a corresponding model operating variable of a model of the parallel hybrid drive, wherein the model does not include the internal combustion engine. The difference between the measured and the calculated size of the model is used as the controller input. A deviation resulting from the comparison of the electrical machine is thus at least partially compensated. A predictive approach is not intended.

The object of the invention is to avoid the disadvantages mentioned and to allow a hybrid vehicle in a simple way a smooth start of the engine - especially while driving. In particular, this should also be possible when using a manual gearbox.

According to the invention this is achieved in that the torque fluctuations occurring during at least one drive wheel entrainment of the engine by means of a powertrain model - preferably continuously - calculated and predicted for the entire Mitschleppvorgang the internal combustion engine, and that the torque fluctuations occurring on the basis of the predicted torque fluctuations active by at least an opposing torque is at least reduced, preferably eliminated.

Preferably, the opposing torque is applied by at least one electric machine.

The powertrain model, preferably based on a two- or multi-mass model, uses as input variables the torques of the electric machine, the internal combustion engine and / or the clutch torque and calculates as output a predicted differential speed between the electric machine and at least one drive wheel of the vehicle, with possible speed ratios between electrical machine and the drive wheel are taken into account. The differential speed forms a regulator input of an anti-jerk controller whose controller output provides the reverse torque.

The clutch torque is modeled according to the slipping or non-slipping clutch condition.

To reduce driveline model inaccuracies, the model output may also be delayed by a defined prediction horizon and compared to a measured magnitude, and error correction of the powertrain model may be performed based on the deviation. The prediction horizon is a predefined value for the duration of the drive train movement. The forecast horizon depends on the particular case, in particular depending on the available computing capacity.

The model output is delayed due to dead times prevailing in the system, for example by communication. These dead times must be known. Then the model size is delayed with the dead time, so that the model size can be compared with the true measured size. Due to this difference, the model is adapted.

In order to prevent a reduction in the torque at the drive wheel during the starting process, it is particularly advantageous if the clutch torque of the clutch is predicted during starting of the internal combustion engine, as long as the separating clutch is in slip, and that on the basis of the predicted clutch torque, the torque of electric machine is pre-controlled. In particular, the torque of the electric machine is increased by the predicted maximum clutch torque of the separating clutch.

With the method according to the invention, very small starting times of the internal combustion engine can be achieved - measured between initiation until the time the drive is taken over by the internal combustion engine become. Due to the implicit error correction, the method has a high robustness against disturbance variables.

This torsional vibrations in the drive train can be largely reduced or even avoided.

In order to start the internal combustion engine during a purely electric drive, the separating clutch, which is located between the internal combustion engine and the electric machine, is closed in a defined manner. The closing process of the separating clutch can be subdivided into the following three phases:

First phase (clutch pulse): In the first phase, the clutch is closed pulsed and partially reopened. During this time, the internal combustion engine is to be accelerated to an ignitable speed (about 300 rpm). For this to happen, the transmitted clutch torque must be greater than the drag torque of the internal combustion engine. Basically, the higher the transmitted clutch torque, the faster the engine reaches an ignitable speed.

Second phase (slipping clutch / speed synchronization): In the second phase, the separating clutch is operated in the slip until the speed of the internal combustion engine has reached approximately the speed of the electric machine. The clutch remains partially closed to speed up the synchronization process,

Third phase (complete closing of the separating clutch): If the differential speed between the internal combustion engine and the electric machine is less than or equal to an applicable parameter, the separating clutch is completely closed. So that the wheel torque in this period corresponds to the driver's request and no / hardly torsional vibrations are induced in the drive train, a control / regulation concept is necessary.

The invention thus has the following aspects:

1.) Pilot control of the electric machine

The clutch torque of the clutch is controlled by the electric machine, as long as the clutch slips. Background: When the clutch is closed and slipping, the clutch transmits its maximum torque depending on its closing force in the direction of negative speed gradient, ie in the direction of the internal combustion engine. If this is not counteracted, the wheel torque would decrease.
Remedy: Pilot control of the clutch torque via the electric motor.

Approach: If the clutch characteristic and the transfer function of the disconnect clutch are known, the transmitted clutch torque can be predicted as long as the clutch is in slip. If you know the clutch torque, you can pilot this with the help of the electric machine. This ensures that the wheel torque corresponds to the driver's request.

Inaccuracies in the clutch model (clutch characteristic + transfer function) and wear of the clutch can cause the electric machine to control a wrong torque. This would lead to driveline vibrations. To counteract this effect, a predictive anti-jerk control is active parallel to the pre-control of the clutch torque.

This can be used in the first and second phases of the starting process.

2.) Predictive anti-jerk control

The predictive anti-jerk control is used to avoid / reduce longitudinal vibrations of the vehicle. Background: When the torque in the powertrain changes with high gradient, torsional vibrations are induced in the driveline. The drive motor oscillates against the reduced mass inertia of the wheel and body. These torsional vibrations manifest themselves to the vehicle occupants in longitudinal vibrations of the vehicle.
Remedy: Predictive anti-jerk control

Approach: An indicator of the jerk is the differential speed between drive wheel and drive motor (electric machine). More precisely, the difference speed is proportional to the jerk. Thus, it makes sense to use the differential speed as a controller input. So that the torsional vibrations can be eliminated as early as possible in the approach, offers a predictive control. To realize this, it is necessary to include a model of the drive train in the control unit. This model is a two or more mass oscillator. Depending on the complexity, this model uses as input variables: the torque of the electric machine, the internal combustion engine and / or the clutch torque and delivers as output the estimated differential speed between the electric motor and the wheel. If all inputs and states of the model are known at the time k = n, then For example, the movement (differential speed) of the drive train can be predicted for the time k = n + j (j: prediction horizon). This predicted movement is used as the controller input.

Thus, with the help of the model, the jerk can be greatly reduced as it is predicted. This opposing torque is provided by the electric machine because it has a very fast response. However, it is also conceivable to apply the counter-directed torque via the internal combustion engine.

Depending on the model (two or more mass oscillators), the currently transmitted disconnect clutch torque is required as input for the model. If this quantity is not provided by any of the controllers (e.g., the transmission controller), this quantity may be calculated via one of the state or output variables of the mass modifier model. For this purpose, one of the angular speeds (internal combustion engine, electric machine, drive wheel) is used and differentiated. By simple equations of motion then the disconnect torque can be estimated. This estimated disconnect torque is then used as input to the powertrain motion model.

The powertrain model is preferably linear to reduce complexity and computational effort. The powertrain itself is nonlinear, however. In addition, the estimated clutch torque of the disconnect clutch is used as an input to the powertrain model. In order to reduce the model inaccuracies, an adaptation of the drive train model is made. For this purpose, the model output is delayed by the prediction horizon (j * sampling time) and then compared with the measured quantity. The powertrain model is adapted via this calculated model error.

The predictive anti-jerk control with adaptation of the powertrain model can be used in the first, second and / or third phases of the starting process.

The invention will be explained in more detail with reference to the figures.

Show it

1 schematically a drive train for carrying out the method according to the invention,

2a the qualitative course without regulation of the torque of the internal combustion engine and the electric machine during a pulse start,

2 B the course of the maximum transferable clutch torque of the clutch during a pulse start,

2c the course of the angular velocity of the internal combustion engine and the electric machine during a pulse start,

3a the simulated curves of the torques of the internal combustion engine and of the electric machine, as well as the clutch torque of the separating clutch during a pulse start, without anti-jerk regulation or control,

3b the simulated course of the vehicle longitudinal acceleration during a pulse start, without anti-jerk regulation or control,

3c the simulated progressions of the rotational speeds of the internal combustion engine and of the electric machine during a pulse start, without anti-jerk regulation or control,

4 a model of the drive train,

5 the overall structure of the scheme of hybrid powertrain and

6a a comparison of the simulated courses of the torques of the electrical machine during a pulse start, with and without anti-jerk regulation or control,

6b a comparison of the simulated courses of the vehicle longitudinal accelerations during a pulse start, with and without anti-jerk regulation or control,

6c a comparison of the simulated progressions of the rotational speeds of the electric machine during a pulse start, with and without anti-jerk control or.

1 shows a parallel hybrid powertrain 8th a vehicle with an internal combustion engine 1 , a separating clutch 2 , an electric machine 3 , a start-up clutch 4 , a gearbox 5 , and a differential 6 which is on drive wheels 7 acts. In purely electric drive of the vehicle by the electric machine 3 is the disconnect clutch 2 opened and the internal combustion engine 1 disabled.

• Erste Phase I: (Kupplungsimpuls): In der ersten Phase I wird die Trennkupplung 2 impulsartig geschlossen und danach teilweise wieder geöffnet. In dieser Zeit soll die Brennkraftmaschine 1 auf eine zündfähige Drehzahl (ca. 300 U/min) bzw Winkelgeschwindigkeit ω_Z beschleunigt werden. Damit dies geschehen kann, muss das übertragene Kupplungsmoment M_TK größer als das Schleppmoment M_VK der Brennkraftmaschine 1 sein. Grundsätzlich gilt: Je höher das übertragene Kupplungsmoment M_TK, desto schneller erreicht die Brennkraftmaschine 1 eine zündfähige Drehzahl.
• Zweite Phase II: (schlupfende Trennkupplung 2/ Drehzahlsynchronisation): In der zweiten Phase II wird die Trennkupplung 2 im Schlupf betrieben, bis die Drehzahl der Brennkraftmaschine 1 annähernd die Drehzahl der elektrischen Maschine 3 erreicht hat. Die Trennkupplung 2 bleibt dabei teilweise geschlossen, um den Synchronisationsvorgang zu beschleunigen,
• Dritte Phase III (vollständiges Schließen der Trennkupplung 2): Ist die Differenzdrehzahl zwischen der Brennkraftmaschine 1 und der elektrischen Maschine 3 kleiner oder gleich einem applizierbaren Parameter, wird die Trennkupplung 2 vollständig geschlossen. Damit das Raddrehmoment in diesem Zeitraum dem vom fahrer vorgegebenen Wunschantriebsdrehmomentes M_F entspricht und keine/kaum Torsionsschwingungen in den Triebsstrang induziert werden, ist ein Steuerungs-/Regelungskonzept notwendig.

Will the internal combustion engine 1 through the electric machine 3 For example, started during a purely electrically driven ride, the starting process takes place in the following three phases I, II, III, as in 2 shown is:

• First Phase I: (Coupling pulse): In the first phase I, the separating clutch 2 closed impulsively and then partially reopened. In this time, the internal combustion engine 1 be accelerated to an ignitable speed (about 300 rev / min) or angular velocity ω _Z. For this to happen, the transmitted clutch torque M _TK must be greater than the drag torque M _{VK of} the internal combustion engine 1 be. Basically, the higher the transmitted clutch torque M _TK , the faster the internal combustion engine reaches 1 an ignitable speed.
• Second phase II: (slipping separating clutch 2 / Speed synchronization): In the second phase II, the separating clutch 2 operated in the slip until the speed of the internal combustion engine 1 almost the speed of the electric machine 3 has reached. The separating clutch 2 remains partially closed to speed up the synchronization process,
• Third phase III (complete closing of the separating clutch 2 ): Is the differential speed between the engine 1 and the electric machine 3 less than or equal to an applicable parameter, the disconnect clutch 2 completely closed. In order for the wheel torque in this period to correspond to the desired drive torque M _F specified by the driver and for no / hardly torsional vibrations to be induced in the drive train, a control / regulation concept is necessary.

In 3 a hybrid start is simulated without anti-jerk control, with In 3a the torque M _{VM of} the internal combustion engine 1 , the torque M _{EM of} the electric machine 3 and the clutch torque M _{TK of} the disconnect clutch 2 are plotted over time t. In 3b is the course of the vehicle longitudinal acceleration a and in 3c the curves of the speed n _{VM of} the internal combustion engine 1 , the electric machine speed n _EM 3 , and the translation ratio adjusted speed n _wheel .

wobei mit

φ_VM

der Drehwinkel der Kurbelwelle der Brennkraftmaschine 1

ω_VM

die Winkelgeschwindigkeit der Kurbelwelle der Brennkraftmaschine 1

J_VM

das Massenträgheitsmoment der Brennkraftmaschine 1

φ_EM

der Drehwinkel des Rotors der elektrischen Maschine 3

ω_EM

die Winkelgeschwindigkeit des Rotors der elektrischen Maschine 3

J_EM

das Massenträgheitsmoment der elektrischen Maschine 3

φ_Rad

der Drehwinkel eines Antriebsrades 7 des Fahrzeuges

ω_Rad

die Winkelgeschwindigkeit eines Antriebsrades 7 des Fahrzeuges

J₂

das Massenträgheitsmoment eines Antriebsrades 7 des Fahrzeuges

i_g

das reduzierte Massenträgheitsmoment des Getriebes 5

c

eine erste Federkonstante des Antriebsstranges 8

d

eine erste Dämpfungskonstante des Antriebsstranges 8

c₂

eine zweite Federkonstante des Antriebsstranges 8

d₂

eine zweite Dämpfungskonstante des Antriebsstranges 8

x ₀

die Anregung in x-Richtung (Fahrzeuglängsrichtung)

y

die Anregung in einer y-Richtung (Querrichtung zur Fahrzeuglängsachse)

bezeichnet ist. 4 shows a replacement model 9 (Powertrain model) of the drive train 8th which is based on the following state space model of equations of motion:

being with

φ _VM

the angle of rotation of the crankshaft of the internal combustion engine 1

ω _VM

the angular velocity of the crankshaft of the internal combustion engine 1

J _VM

the mass moment of inertia of the internal combustion engine 1

φ _EM

the angle of rotation of the rotor of the electric machine 3

ω _EM

the angular velocity of the rotor of the electric machine 3

J _EM

the mass moment of inertia of the electric machine 3

φ _wheel

the angle of rotation of a drive wheel 7 of the vehicle

ω _wheel

the angular velocity of a drive wheel 7 of the vehicle

J ₂

the moment of inertia of a drive wheel 7 of the vehicle

i _g

the reduced mass moment of inertia of the gearbox 5

c

a first spring constant of the drive train 8th

d

a first damping constant of the drive train 8th

c ₂

a second spring constant of the drive train 8th

d ₂

a second damping constant of the drive train 8th

x ₀

the excitation in the x-direction (vehicle longitudinal direction)

y

excitation in a y-direction (transverse to the vehicle's longitudinal axis)

is designated.

In 5 is the overall structure of the scheme shown schematically. By the driver 10 a desired drive torque M _{F is} specified. It comes in 11 to a division of the desired drive torque M _F into a drive torque M _{VM of} the internal combustion engine 1 and in a driving torque M _{EM of} the electric machine 3 ,

The inventive method provides two mechanisms to enable a smooth start of the internal combustion engine: feedforward control and anti-jerk control.

1.) pilot control

At a pulse start of the internal combustion engine, the clutch torque M _{TK of} the separating clutch 2 over the electric machine 3 pilot operated, as long as the separating clutch 2 slips. When the disconnect clutch 2 is closed and the internal combustion engine 1 can still deliver torque M _VM (since not yet ignited), transmits the clutch 2 their maximum torque M _TK (depending on the closing force) in the direction of the internal combustion engine 1 , If not counteracted, the torque M _{wheel of} the drive wheel 7 reduce. To avoid this, the clutch torque M _TK on the electric machine 3 piloted by using the feedforward control 15 a pre-control torque M _{TKV is} requested. This can - if the clutch characteristic and the transfer function of the separating clutch 2 are known - done by the fact that the transmitted clutch torque M _TK the disconnect clutch 2 is predicted, as long as the disconnect clutch 2 in slippage. Does one know the clutch torque M _{TK of} the clutch 2 , you can do this with the help of the electric machine 3 forward system. This ensures that the _wheel torque M _Rad approximately corresponds to the driver's desired torque M _F.

Inaccuracies in the clutch model 12 (Clutch characteristic + transfer function) and signs of wear on the disconnect clutch 2 However, this can cause the electric machine 3 a wrong torque pre-steers. This would lead to driveline vibrations. To counteract this effect, is parallel to the pilot control of the clutch torque 2 a predictive anti-jerk control 13 active. The anti-jerk control 13 uses the mentioned powertrain model 9 , as well as a mathematical coupling model 12 for calculating the clutch torque M _{TK of} the disconnect clutch 2 For example, a two-or multi-mass model.

The pilot control of the clutch torque can be used in the coupling phases 1 and 2.

2.) Predictive anti-jerk control

The predictive anti-jerk control is performed to avoid or prevent longitudinal vibrations of the vehicle along the longitudinal axis x. The torques in the drive train change 8th with high gradient, torsional vibrations are in the drive train 8th induced. The drive motor vibrates (electric machine 3 ) against the reduced mass inertia of drive wheel 7 and bodywork. These torsional vibrations manifest themselves to the vehicle occupants in longitudinal vibrations (jerking) of the vehicle.

An indicator of the jerk is the differential speed Δn EM- _{> wheel} between the drive wheel and the electric machine 3 , wherein the Differnzdrehzahl .DELTA.n EM- _{> Rad is} proportional to the jerk. Thus, it makes sense the differential speed Δn EM- _{> Rad} as a controller input size for the anti-jerk controller 13 to use. So that the torsional vibrations can be eliminated as early as possible in the approach, a predictive control is performed in the inventive method. To realize this, it is necessary a powertrain model 9 in the control unit 14 counted. This powertrain model 9 is essentially a two- or multi-mass oscillator and uses - depending on the complexity - as input variables, the torque M _{EM of} the electric machine 3 , the torque M _{VM of} the internal combustion engine 1 , and / or the clutch torque M _{TK of} the disconnect clutch TK and provides as output the - overall gear ratio adjusted - estimated predicted differential speed Δn EM- _{> Rad} or predicted differential angular velocity Δω EM- _{> Rad} between electric machine 3 and drive wheel 7 , Are all input variables and states of the drive model 9 at time k = n, the movement (predicted differential speed Δn EM- _{> wheel} or predicted differential angular speed Δω EM- _{> wheel} ) of the drive train can be determined 8th are predicted for the time k = n + j (j: prediction horizon). This predicted movement (predicted differential rotational speed Δn EM- _{> Rad} or predicted differential angular velocity Δω EM- _{> Rad} ) is used as the regulator input for the anti-jerk control 13 used.

Thus, with the help of the powertrain model 9 the jerk will be greatly reduced as it is predicted. This opposing correction _torque M _{Korr is presented} via the electric machine 3 because she has a very fast response. But it is also conceivable about the internal combustion engine 1 to apply the counter-directed correction _torque M _Korr .

Depending on the powertrain model 9 (Two- or multi-mass oscillator) is the currently transmitted clutch torque M _TK the disconnect clutch 2 as input for the powertrain model 9 needed. If this variable is not supplied by one of the control units (eg the gearbox control unit), this variable can be via one of the state or output variables of the clutch model formed by a mass-swinging model 12 be calculated. For this purpose, one of the angular velocities ω _VM , ω _EM , ω _{wheel of} the internal combustion engine 1 , the electric machine 3 or the drive wheel 7 used and differentiated. About equations of motion can then the clutch torque M _{TK of} the clutch 2 to be appreciated. This estimated clutch torque M _{TK of} the disconnect clutch 2 is then used as input for the powertrain model 9 used.

The powertrain model 9 is linear. The powertrain 8th in itself, however, is nonlinear. In addition, the estimated clutch torque M _{TK of} the disconnect clutch 2 as input for the drive model 9 used. To reduce the model inaccuracies, an adaptation of the powertrain model 9 made (Reference: Luenberger observer 16 ). For this purpose, the model output variable (predicted differential rotational speed Δn EM- _{> wheel} or predicted differential angular velocity Δω EM- _{> Rad} ) in the idle timer 17 delayed by the prediction horizon (j · sampling time) and then compared with the measured quantity (actual differential speed Δn EM- _{> wheel, akt} or differential angle speed Δω EM- _{> wheel, akt} ). About this calculated model error e becomes the powertrain model 9 adapted.

The occurring torque fluctuations on the basis of the predicted torque fluctuations can thus be actively reduced by at least one opposing torque at least, preferably eliminated.

The predicted anti-jerk control with adaptation of the powertrain model 9 can be used in phases I, II, and / or III.

In the case of a pulse start of the internal combustion engine 1 through the electric machine 3 be the entrainment of the internal combustion engine 1 occurring torque fluctuations by means of the powertrain model 9 calculated and for the entire Mitschleppvorgang the internal combustion engine 1 predicted. The powertrain model 9 uses as input variables the torque M _{EM of} the electric machine 3 and the clutch torque M _{TK of} the disconnect clutch 2 , wherein the clutch torque M _TK . the separating clutch 2 by means of a coupling model 12 on the basis of the torque M _{EM of} the electric machine 3 and the one by the powertrain model 9 calculated longitudinal excitation x _k + j is calculated at the time k + j. As an output, says the powertrain model 9 a - in terms of the overall ratio between electrical machine 3 and drive wheel 7 adjusted - predicted differential speed Δn EM- _{> Rad} or a predicted differential angular velocity Δω EM- _{> wheel} between the electric machine 3 and at least one drive wheel 7 , The predicted differential speed Δn EM- _{> Rad} or the predicted differential angular speed Δω EM- _{> Rad} becomes the controller 13 fed, which calculates therefrom a correction torque M _Korr , which in the entrainment of the internal combustion engine 1 occurring torque fluctuations is opposite.

To reduce inaccuracies of the powertrain model 9 is the model output predicted differential speed .DELTA.n EM- _{> Rad} or a predicted Differenzwinkelgeschwindigkeit .DELTA.ω EM- _{> Rad} delayed by a defined prediction horizon j and predicted with a measured magnitude differential speed Δn EM- _{> Rad} or a predicted differential angular velocity .DELTA.ω EM- _{> Rad, akt} compared and based on the deviation, an error correction of the powertrain model 9 performed.

Furthermore, a reduction of the torque M _Rad on the drive wheel 7 be prevented during the starting process when the clutch torque M _TK the disconnect clutch 2 during starting of the internal combustion engine 1 is predicted, as long as the disconnect clutch 2 is in slip, and that on the basis of the predicted clutch torque M _TK the torque of the electric machine 3 is precontrolled, for example, the torque M _{EM of} the electric machine 3 around the predicted maximum clutch torque M _{TK of} the disconnect clutch 2 is increased. This can reduce the torque on the drive wheel 7 by torque drain to the engine 1 be compensated.

In 6 is a pulse start with anti-jerk control 13 a pulse start without anti-jerk regulation or control 14 juxtaposed, each in 6a the torque M _{EM of} the electric machine 3 , in 6b the vehicle longitudinal acceleration a and in 6c the speed n _{EM of} the electric machine 3 is plotted over time t. The dotted line shows the course without anti-jerk control, the solid line with predictive anti-jerk control. It is clear in 6b to see that the vehicle longitudinal acceleration a with the inventive predictive anti-jerk control 13 can be significantly reduced.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102006034937 A1 [0003]
• DE 102011002742 A1 [0003]
• DE 102011109353 A1 [0005]
• DE 19814402 A1 [0006]
• DE 102006047655 A1 [0007]

Claims (8)

Method for starting an internal combustion engine ( 1 ) of a hybrid vehicle during at least one purely electric drive mode, wherein in at least one phase of the starting one between the internal combustion engine ( 1 ) and an electric machine ( 3 ) arranged separating clutch ( 2 ) at least partially closed and the internal combustion engine ( 1 ) by the electric machine ( 3 ) is towed, characterized in that the entrainment of the internal combustion engine ( 1 ) on at least one drive wheel ( 7 ) occurring torque fluctuations by means of a drive train model ( 9 ) - preferably continuously - calculated and for the entire Mitschleppvorgang the internal combustion engine ( 1 ), and that the occurring torque fluctuations based on the predicted torque fluctuations are actively at least reduced, preferably eliminated, by at least one opposing correction torque (M _Korr ). A method according to claim 1, characterized in that the opposing torque of at least one electric machine ( 3 ) is applied. A method according to claim 1 or 2, characterized in that - preferably based on a two- or multi-mass model - - powertrain model ( 9 ) as an input variable at least one variable from the group torque (M _EM , M _VM ) of the electric machine ( 3 ), Torque (M _VM ) of the internal combustion engine ( 1 ) and coupling torque (M _TK ) of the separating clutch ( 2 ) and as an output a predicted differential speed (Δn EM- _{> Rad} ) or a predicted differential angular velocity (Δω EM- _{> Rad} ) between the electric machine ( 3 ) and at least one drive wheel ( 7 ) of the vehicle. Method according to Claim 3, characterized in that the predicted differential rotational speed (Δn EM- _{> Rad} ) or the predicted differential angular velocity (Δω EM- _{> Rad} ) comprises a regulator input variable of an anti-jerk regulator ( 13 ) whose controller _output provides the _reverse correction _torque (M _Korr ). A method according to claim 3 or 4, characterized in that the coupling torque (M _TK ) of the separating clutch ( 2 ) by means of a mathematical coupling model ( 12 ), preferably a two or more mass model. Method according to one of claims 1 to 5, characterized in that to reduce inaccuracies of the powertrain model ( 9 ) the model output is delayed by a defined prediction horizon (j) and compared to a measured quantity, and based on the deviation, an error correction of the driveline model ( 9 ) is made. Method according to one of claims 1 to 6, characterized in that the clutch torque (M _TK ) of the separating clutch ( 2 ) during starting of the internal combustion engine ( 1 ) is predicted, as long as the separating clutch ( 2 ) is in slip, and that on the basis of the predicted clutch torque (M _TK ) the torque (M _EM ) of the electric machine ( 2 ) is piloted. Method according to one of claims 1 to 7, characterized in that in a first phase (I) between the internal combustion engine ( 1 ) and at least one electric machine ( 3 ) arranged separating clutch ( 2 ) is closed pulse-shaped and partially reopened, in a second phase (II) the separating clutch ( 2 ) is operated in the slip until the speed (n _VM ) of the internal combustion engine ( 1 ) at least approximately the rotational speed (n _EM ) of the electric machine ( 3 ), and in a third phase (III) the separating clutch ( 2 ) is completely closed.

Images