DE102016215787A1 - Method for adapting torque coordination when restarting an internal combustion engine in a hybrid vehicle - Google Patents

Abstract

Method comprising evaluating at least one compensated restart of an internal combustion engine (1) by detecting the time behavior between the clutch torque (MK0) of a separating clutch (5) and the compensating torque (MKomp) provided by an electric machine (2) by evaluating the drive rotational speed (ωEM) and / or the speed gradient (ω, EM) of the electric machine (2) during the construction of the compensation torque (MKomp) and / or the clutch torque (MKup).

Description

The invention relates to a method for operating a hybrid vehicle, in particular for the adaptation of torque coordination when restarting an internal combustion engine in a hybrid vehicle.

The start of the internal combustion engine takes place during the electric driving operation of a hybrid vehicle by coupling the internal combustion engine to the drive train, which is driven by an electric motor. A necessity for starting the internal combustion engine results, for example, when the driver requests more torque than can be provided by the electric motor as the drive torque, or when the energy content of an energy store, which supplies the electric motor with energy, has dropped too much.

From the international patent application WO 2011/064018 For example, a method for operating a hybrid vehicle is known, in which the internal combustion engine is started by a part of the drive torque applied by the electric motor by bringing a separating clutch arranged between the first and the second drive unit from an open to a slipping state. For the start an additional compensation torque is provided by the electric machine.

From the German patent application DE 10 2008 004 366 A1 is a drive train arrangement of a vehicle with an internal combustion engine is known, which can be started on the basis of a controllable by means of a clutch and controllable by means of a speed controller electric machine. It is provided that a detection of a control value of the speed controller towards the end of at least one startup operation of the internal combustion engine is still under slip control underlying coupling and a comparison of the control value with a default value is used to determine a value deviation, which can be compensated by means of a correction value in a reboot is. A similar technique is also from the German patent application DE 10 2011 084 844 A1 known.

During the re-start of the internal combustion engine as possible, the moment balance, which results from a free cut of the electric motor should be satisfied at all times, ie the moments around the electric motor should cancel out in the sum. For this purpose, the actually transmitted from the clutch torque should be known. For this purpose, there are methods for the adaptation of a clutch / clutch characteristic, such as that in the published patent application DE 10 2008 004 366 A1 revealed. In addition, the moment actually transmitted by the disconnect clutch and the compensation torque of the electric machine should be set simultaneously and be equal in magnitude. It all comes first on the clutch to delay between Sollmomentenvorgabe and actual torque. These are due to the inertia of the disconnect path and physical effects of the disconnect clutch.

The object of the present invention is to provide a method for operating a hybrid vehicle which at least partially overcomes the above-mentioned problems.

This object is achieved by the method according to claim 1. Further advantageous embodiments of the invention will become apparent from the subclaims and the following description of preferred embodiments of the present invention.

The method according to the invention comprises evaluating at least one compensated restart of an internal combustion engine by detecting the time behavior between the clutch torque of a disconnect clutch and the compensation torque provided by an electrical machine by evaluating the drive speed and / or the rotational speed gradient of the electric machine during the construction of the compensation torque and / or clutch torque.

In this way, a bad timing can cause, for example, a fall in the rotational speed of the electric machine or a turning out of the rotational speed of the electric machine during a restart. The present method can detect good or bad timing. In particular, changes in the behavior of the clutch with respect to the temporal behavior for the conversion of desired to actual torque during the compensated slip start are detected and can be learned again. Also, insufficient timing, which leads to loss of comfort, triggered by different behavior between the clutches in different transmissions, can be reduced in the long term.

The method is suitable for example for plug-in hybrid vehicles with dual-clutch transmission, which contain a controllable disconnect clutch and to start the internal combustion engine perform compensated slip starts. The vehicle can use, for example, an electro-hydraulically actuated multi-plate clutch (wet clutch) or a dry disconnect clutch.

In a preferred embodiment of the method, the rotational speed of the electric machine is observed and evaluated as an offset to the rotational speed of an active transmission input shaft. If the speed of the electric machine is related to the speed of the active transmission input shaft, it can be recognized in particular if the speed of the electric machine falls below the input speed of the transmission, since the excess torque of the separating clutch is no longer available to the output. This should be avoided, as it comes in this case to a so-called train-thrust change and thus to a noticeable jerk in the drive train.

In a further preferred embodiment, the method comprises adapting the time behavior between the clutch torque and the compensation torque. The adaptation of the time behavior between the clutch torque and the compensation torque can be based, for example, on a weighting function.

In a further preferred embodiment, the method comprises postadapting the time constant of the compensated restart. The time constants may be defined, for example, as a function of transmission temperature and setpoint pressure. The Nachadaptieren (re-learning) of the time constant can be done for example by adjusting time constants. The time constants are usually applied and stored in characteristic diagrams (eg PT1 and / or PT2 characteristic diagrams) of the transmission control and can be adapted by the method according to the invention. An advantage of this is, for example, that a standard specification could be delivered with regard to the timing and the timing optimally adjusted by learning. This is particularly advantageous over firmly applied solutions that can not respond to lifetime effects.

In a further preferred embodiment, noticeable re-starts are detected via a counter, and post-adaptation is performed only after a certain number of noticeable re-starts. For example, the counter counts up if the quality rating of a restart is worse than a predefined quality value. Thus it can be avoided that short-term effects are directly re-learned and negatively influence later re-starts. Rather, permanent and reproducible system changes over mileage are compensated and adapted.

In a further preferred embodiment, the method comprises evaluating the driving situation on the basis of boundary conditions (ambient conditions) and / or a current drive train situation, a transition between active and passive state of a zero-start evaluation taking place on the basis of this evaluation.

In a further preferred embodiment, the method comprises recognizing defined phases of a restart. The recognition of the defined phases of a restart can, for example, take place via an internal functional status. The recognition of defined phases of a restart, for example, relevant to the o. G. Identification of influences that are not due to the time behavior between the clutch torque of a separating clutch and the compensating torque provided by an electric machine.

In a further preferred embodiment, the method comprises detecting the speed gradient of the internal combustion engine.

In a further preferred embodiment, the method comprises calculating the transmitted clutch torque via the speed gradient of the internal combustion engine and a plausibility check of the clutch characteristic curve. Thus, the method is able to identify other causes for a break-in or unscrewing the drive speed and possibly exclude or exclude a bad timing as a cause.

In a further preferred embodiment, the method comprises evaluating the change in the driver's desired torque and other environmental conditions during a restart.

The method according to the invention can be implemented as a computer-implemented method in, for example, a transmission control module having a processor, memory and communication interfaces. In particular, the method may be in the form of software program instructions implemented on a processor, e.g. B. a transmission control module, running. The invention is therefore also directed to a processor adapted to implement the method described, or to controllers comprising a processor arranged in this way. Also, a motor vehicle is subject of this invention, which includes such control.

Embodiments of the invention will now be described by way of example and with reference to the accompanying drawings, in which:

1 shows a schematic and exemplary schematic representation of a torque adding parallel hybrid drive concept;

2 shows the EM compensation torque, the target K0 torque, as well as speeds of an electric machine, an internal combustion engine, and a transmission input shaft during different phases of a restart;

3 shows an exemplary embodiment of a post-learning process; and

4 show further exemplary processes that performs a control to realize an adaptation of the torque coordination at restart.

PARALLEL HYBRID DRIVE CONCEPT

1 provides a schematic and exemplary schematic representation of a torque adding parallel hybrid drive concept. The drive concept may be, for example, a plug-in hybrid powertrain having a single-shaft parallel hybrid drive train or the like. A parallel hybrid powertrain of a hybrid vehicle includes an internal combustion engine 1 (VKM), a dual-mass flywheel 8th (ZMS), an electric machine 2 (EM), a gearbox 3 and a differential 4 , Between gears 3 and internal combustion engine 1 There is a controllable separating clutch 5 (K0). The gear 3 is designed for example as an electro-hydraulically actuated dual-clutch transmission. A high-voltage battery 6 provides electrical energy to power electronics 7 available, which allows the electric machine 2 operates. With the help of the separating clutch 5 can the hybrid vehicle, inter alia, in the operating modes "electric driving" and "recuperation" are added. Furthermore, the internal combustion engine 1 can be operated at full power and in addition, the electric machine 2 provide a torque to increase the total torque (boosting).

In the drive train, the internal combustion engine 1 while driving via the adjustable separating clutch 5 (K0) - also called hybrid clutch or Wiederstartkupplung - started by this slipping operated. The desired K0 torque (for starting the internal combustion engine 1 ) is in the electric machine 2 held as a Nederstar reserves and during the start of the electric machine 2 in addition to the driving torque provided. The electric machine 2 Thus "compensates" the additionally acting on the driveline disconnect clutch torque. This is intended to delay the output due to the addition of the electric machine 2 acting load (K0) can be prevented. The restart thus takes place by means of compensated slip starts.

In order that the torque balance (sum of all moments within the balance envelope around the EM = 0 Nm) is valid, this should actually be done by the separating clutch 5 transmitted torque and the compensation torque of the electric machine 2 be synchronized and equal in terms of amount. This leads to delay times between the setpoint torque input and the actual actual torque (significant for the disconnect clutch 5 due to the inertia of the disconnect path and physical effects of the disconnect clutch). Both the timing between the K0 actual torque and the EM compensation torque, as well as the respective gradients are coordinated with each other. The application is based on metrologically determined values and delay times and can be configured via temperature and control pressure. However, the behavior, primarily the separating clutch, can change in the long run on mileage, oil cleanliness, etc. Even manufacturing tolerances lead to different coupling behavior. In these cases, the application of the timing between K0 torque and EM compensation torque no longer match. This can cause two effects. An unfavorable timing can cause a fall in the rotational speed of the electric machine or a turning out of the rotational speed of the electric machine during a restart.

When breaking the speed of the electric machine, the K0 actual torque is set earlier by the disconnecting clutch / built up faster than the EM compensating torque. As a result, the separating clutch takes more momentum than is compensated by the electric machine. In sum, therefore, all active clutches (K0 and clutches) take more momentum than the electric machine provides. The torque balance over the electric machine is no longer correct. As a result, the EM speed breaks and falls below the input speed of the transmission. It comes to a so-called train-thrust change and thus a noticeable jerk in the drive train.

When unscrewing the speed of the electric machine, the K0 actual torque is provided by the clutch later than the EM compensation torque. In sum, all active clutches (K0 and clutches) take less torque than the electric machine provides. Consequently, the electric machine continues to rotate beyond the output speed. This case is acoustically noticeable.

These described timing problems can occur, for example, via a change in the clutch behavior over life / mileage.

The method described below provides that a constant deviation from the actual torque of the clutch to the desired torque of the clutch, z. B. due to not completely correct clutch characteristic, is identified and is not mistakenly assigned to the timing problem. Ie. a speed response of the drive due to a deviation between the desired torque and actual torque does not lead to Nachadaptieren. For this purpose, the restart is divided into different phases and the speed gradient of the VKM is monitored and evaluated.

MOMENT BALANCE (PHYSICAL BASICS)

The physical basis of the method is given by the moment balance ΣM _EM around the electric machine (EM): ΣM _EM = 0 = M _EM - M _K0 - M _K12 - θ · ω. _EM with _EM M = M + M _{driver's wish Comp} + M _{interventions}

Here, M _EM denotes the torque provided by the electric machine, M _K0 denotes the torque absorbed by the separating clutch, M _Komp denotes the compensation torque, M _{interventions} denotes the torque _engagement with the electric machine, M _K12 denotes the torque of the active driving clutch, M denotes _driver's intention the torque requested by the driver and ω _EM denotes the drive speed of the electric machine and θ the reduced inertia of the electric machine, K0 secondary side and K12 primary side.

This results in the speed gradient of the electric machine according to ω. _EM = 1 / θ · (M _EM -M _K0 -M _K12 )

This applies if the separating clutch K0 and the active driving clutch K12 are slipping during the restart. As a result, both active couplings transmit their coupling capacity according to K _Kup = F _Anpress · μ · r _m · z

With F _pressure = p _hydr · A for wet clutches, the clutch friction coefficient μ, the mean friction radius r _m and the number z of the friction surfaces.

This adjusts the effective clutch torque accordingly M _Kup = sgn (ω _a - ω _out ) · K _Kup one.

unter Vernachlässigung des Zweimassenschwungrades auf der VKM-Seite.The transmitted K0 clutch torque can be plausibilized on the basis of a VKM-side torque balance ΣM _VKM . For the unfired slip start applies here according to

neglecting the dual-mass flywheel on the VKM side.

Here, M _VKM denotes _loss, the loss torque of the internal combustion engine, M _K0 denotes the torque transmitted by the separating clutch, and ω. _VKM denotes the speed gradient of the internal combustion engine and θ _VKM the inertia of the internal combustion engine.

It follows that the actual K0 moment M _K0 by loss torque M _{VKM, loss} and VKM speed gradient ω. Calculate _VKM and compare with the setpoint. The VKM-side torque balance is used here only for plausibility of the K0 actual torque and limits the cause of a poor start.

OBSERVATION OF THE DRIVE SPEED

The embodiment of the invention set forth below provides for monitoring the drive speed during the restart. This speed changes according to the above physical basis, i. H. analogous to the error moment in the balance envelope of the electrical machine.

According to a first embodiment, the absolute EM speed ω _{EM is} observed and evaluated as an offset Δω to the speed of the active transmission input shaft (GEW) ω _GEW : Δω = ω _EM - ω _GEW

The desired value Δω _soll is selected in this embodiment as Δω _soll = 200 min ^-1 . The evaluation of restart-up is carried out in the following embodiment, by comparing the measured offsets Δω _is with the target value _to Δω. In this embodiment, the determined weighting factor B _{Abs is divided} on a scale of -10 to +10 by applying applicable speed thresholds are stored. Exemplary speed thresholds are for the nominal value Δω _to given = 200 min ^-1 in the following table: Δω _should Δω _is B _Abs 200 min ^-1 400 min ^-1 10 200 min ^-1 360 min ^-1 8th 200 min ^-1 320 min ^-1 6 200 min ^-1 280 min ^-1 4 200 min ^-1 240 min ^-1 2 200 min ^-1 200 min ^-1 0 200 min ^-1 160 min ^-1 -2 200 min ^-1 120 min ^-1 -4 200 min ^-1 80 min ^-1 -6 200 min ^-1 40 min ^-1 -8th 200 min ^-1 0 min ^-1 -10

In a manner known to those skilled in the art, threshold value evaluation or interpolation can be performed on the basis of the stored values.

In this example, the evaluation quantity B _Abs is designed so that the evaluation quantity B _Abs = 0 if the offset Δω corresponds to the setpoint value Δω _soll . On the other hand, if the EM speed breaks down so much that it falls below the input speed of the transmission (Δω _is <0), the result is a rating of -10. Thus, a rating of -10 indicates that there is thus a "train-push change".

In the illustrated embodiment, the above evaluation parameter was _Abs B as a function of the measured _offsets Δω and the target value Δω _to specified. It is obvious to the person skilled in the art that the evaluation quantity B _{Abs can} alternatively also be specified directly as function B _Abs (ω _EM , ω _GEW ) of the absolute EM rotational speed ω _EM and the rotational speed of the active transmission input shaft ω _GEW , e.g. B. in the form of a stored in the memory of the controller table with correspondingly predetermined values. Alternatively, other methods could be used to determine the rating size, e.g. The minimization of a cost functional, or the like.

Alternatively or additionally, the speed gradient ω. _{EM of} the electrical machine are observed and evaluated. The speed gradient ω. _EM indicates the amount of the fault torque in the EM balance envelope. According to one embodiment, an evaluation formula is used for this purpose, in each case the EM rotational speed ω _EM and the EM rotational speed gradient ω. _EM weighted with a factor, according to

The factors factor _abs and factor _degree can be applied. The evaluation variable B _Ges determined in this exemplary embodiment is again divided on a scale of -10 to +10 by depositing applicable speed and gradient thresholds.

According to another embodiment, the evaluation is based only on the speed gradient ω. _EM performed by applying applicable gradient thresholds.

According to further embodiments, however, the evaluation can also be carried out by means of theoretical tools of multivariate analysis methods.

The absolute EM speed ω _EM , the EM speed gradient ω. _EM and the rotational speed of the active transmission input shaft ω _GEW , which serve as the starting point of the evaluation in the above embodiments, obtain the control from, for example, the engine control and the transmission control of the vehicle, respectively. They are determined there in a manner known to the person skilled in the art.

ACTIVATION CONDITIONS OF THE EVALUATION FUNCTION

According to one embodiment variant, the evaluation function is activated / deactivated via applicable values of some environmental conditions. The reason for this is the avoidance of unwanted influences that influence the results. The assessed starts are limited as far as possible to stationary and reproducible situations. Thus, the dispersion of the results is reduced.

Applicable activation conditions are u. a. the active gear, the transmission temperature, the current driveline situation, the absolute driver's desired torque and the change of the driver's desired torque during the restart, the absolute drive speed, or the like. For example, at low oil temperatures, the oil viscosity is greater, so that it can be established as an activation condition that an assessment of restarting should be made only from an oil temperature of 30 ° C. At low speeds, the dynamics are often large, so it may be further advantageous to only allow a re-start rating in high gears, e.g. B. in the corridors 3 to 6 or 7 make. Furthermore, it can also be specified as a further activation condition that an evaluation of restarting during switching operations should be dispensed with.

RECOGNITION OF REPEAT PHASE AND INFLUENCE ANALYSIS

The above-described change in the observed input speed may have different causes. Depending on the phase of the restart, several influences may be the cause. For this reason, the restart of the VKM is divided into different phases.

Advantageous for the evaluation function is the detection of when the EM speed changes in the restart. In order to identify the phase in which the EM speed changes and the influence of this behavior, the following three restart phases are defined: Phase 1: clutch fill, Phase 2: VKM break and torque build up, and Phase 3: VKM pull-up. 2 shows the EM compensation torque (M _{EM, Komp} ), the K0 set torque (M _{K0, setpoint} ), as well as speeds of an electrical machine (n _EM ), an internal combustion engine (n _VKM ) and a transmission _input shaft (n _EW ) during these phases of a restart.

The evaluation of a restart is done in phase 2. Phase 1 and 3 serve, among other things, for the identification of phase 2. For the detection of the phases, the momentary gradients are considered, as they are, for example, in 2 are recognizable. In phase 2, a re-start evaluation is made based on the absolute EM speed ω _EM , the EM speed gradient ω. _EM and the speed of the active transmission input shaft ω _GEW , as described in the section "Monitoring the input speed" above.

In the considered phase 2, a timing problem between K0 torque and EM compensation torque may be the cause of an unwanted speed curve of the electric machine. In addition, the K0 characteristic may have an error or the disconnect clutch K0 may have a hardware fault torque (valve clamp, etc.). For this reason, in phase 2 the K0 torque is additionally plausibilized via the VKM-side torque balance described above by observing and evaluating the VKM speed gradient.

From the VKM-side torque balance results (see section "moment balance" above):

It follows that the actual K0 moment M _K0 by loss torque M _{VKM, loss} and VKM speed gradient ω. Calculate _VKM and compare with a setpoint M _K0 . This balance can be used to determine a K0 torque error ΔM _K0 = M _K0 - M _{K0, soll} . The setpoint M _{K0, soll} is delivered, for example, by a transmission control. If the torque ΔM _K0 actually transmitted by the separating clutch is higher / lower than an applied (predetermined) threshold (| ΔM _K0 |> threshold) and thus appreciably above or below the desired torque (M _{K0, soll} ), the timing is excluded as the sole cause ,

In summary, according to this embodiment, the re-start for the analysis of the correct influencing factor is divided into phases (of which only one phase is considered for the evaluation) and additionally the actually acting K0-moment is plausibilized via the VKM-side moment balance.

EVALUATION

In the crucial phase 2, the VKM is touched. For the formation of a quality criterion, both the absolute speed offset from the engine speed to the transmission input speed and the engine speed gradient are observed at this stage, as described in the section "Observation of the Drive Speed" above.

The method may also identify and account for the influence of the active drive clutch on the accounting envelope of the electrical machine. In order to also identify the influence of the active drive clutch on the balance shell of the electrical machine, the error torque of K12 can be determined metrologically in a reference phase with the clutch K0 still fully open.

ADJUSTMENTS / relearning

If the timing of the moments is identified as the cause for the insufficient re-start comfort, this can be readjusted (adapted). If slippage ceases, the electric machine provides too late compensation for the K0 torque. The compensation moment should therefore be set earlier. If the electric machine turns over the output speed, the compensation should be delayed.

The torque coordination between separating clutch and electric machine can be done, for example, by the electric machine is influenced so that it provides a time delay on their electrical torque. This can be done, for example, by the transmission control sending the command to build up the EM compensation torque delayed. The time delay can be realized by means of maps which provide time constants for PT1 elements. For example, the time constants in maps can be defined as a function of transmission temperature and setpoint pressure. According to the method of the invention, the time constants are also dynamically re-learned and post-adapted. The Nachlernen and Nachadaptieren the adaptation can by means of appropriate maps for timing and shaping of the Compensation torque as PT2 element (based on the physical behavior of the K0 actual torque). The PT2 element can be implemented, for example, by two PT1 links connected in series. Instead of permanently applied time constants of the two PT1 characteristic maps, therefore, variable time constants are used according to the invention. Also conceivable is the structure as a linear function, exponential function, or as a PT1 element.

According to one embodiment, to compensate for the correct timing, the compensation is shifted, for example, by a fixed factor (about 1 ... 10 ms). According to an alternative embodiment, to compensate for the correct timing, the compensation is shifted depending on the quality of the start (or the last start). The quality of the start is evaluated, for example, as set forth in the section "Monitoring the input speed" above, that is, for example, based on the evaluation quantities B _Ges , B _Abs , B _degrees , or the like.

According to a preferred embodiment, the evaluation function is designed to be observationally active with as many fully compensated starts as possible and to detect a significant change in the restarting quality. According to this embodiment, it is not readjusted in the sense of a rapid evaluation function after each restart, but is detected via a counter, the "objected" restart and read only after a certain number (applicative). The counter counts up if the quality rating of a startup is worse than an applied (predefined) quality value.

Thus it can be avoided that short-term effects are directly re-learned and negatively influence later re-starts. Rather, permanent and reproducible system changes over mileage should be compensated and adapted. The evaluation function is not intended to implement any short-term changes, but to successively re-learn the influences on mileage. Specifically, a learning process, for example, should not take place on the basis of a single start. Thus, the evaluation function is designed so that at least Y re-starts (variable Y) with a rating of at least X (variable X) are necessary before they are recognized as reproducible burglaries and an adaptation takes place. In this context, it is also ensured that different types of abnormalities in alternation do not require adaptation.

3 shows an embodiment of a post-learning process. At step 301 , a counter c is initialized with the starting value zero (c = 0). Then the procedure goes to step 302 continued. At step 302 a restart (WS) is detected, for example by a corresponding start request from the engine control unit to the transmission control unit via CAN is detected. The corresponding CAN signal is in step 302 evaluated. If a restart is detected, then the method goes to step 303 continued. At step 303 the restart is evaluated to determine a score B _tot , for example by applying the evaluation method described above. At step 304 is the rating made plausible (PL?). If it is determined in the plausibility check that (N) the score B _{tot is} not plausible (the timing of the compensation is not responsible for the rating), then the method returns to step 302 back to recognize the next reboot and make it plausible. Is determined at the plausibility that (Y) the evaluation is B _ges plausible (the timing of the compensation is responsible for a review), then the method proceeds to step 305 continued. At step 305 it is checked whether the score B _{ges is} greater than or equal to a predefined value + X (B _ges ≥ + X?). If this is the case (Y), the method goes to step 306 where counter c is incremented by one (c = c + 1). Otherwise (N), the method goes to step 307 continued. At step 307 it is checked whether the score B _{ges is} less than or equal to a predefined value -X (B _ges ≦ -X). If this is the case (Y), the method goes to step 308 where counter c is decremented by one (c = c - 1). At step 309 it is then checked whether the counter has exceeded a predefined value + Y (c ≥ + Y). If this is the case (Y), then the method goes to step 310 where re-learning takes place with respect to a "speed increase" (+ ω). Otherwise (N), the method goes to step 311 continued. At step 311 it is checked whether the counter has fallen below a predefined value -Y (c ≦ -Y). If this is the case (Y), then the method goes to step 312 where an adaptation to a "speed dip" occurs (-ω). Otherwise (N), the process returns to step 302 back to see a new restart and to repeat the process. After making an adaptation to a "Speed Rise" or "Speed Drop" in the steps 310 respectively. 312 the procedure returns to step 301 back where the counter c is reset to zero and the described process is executed again.

In alternative embodiments, an additional filtering of the drive speed and the Nachlernfaktoren / Verstellfaktoren can be provided.

CONSTRUCTION AND STRUCTURE OF THE EVALUATION FUNCTION

4 shows another embodiment of processes that performs a control to implement an adaptation of the torque coordination at restart.

In a first step 401 the controller evaluates the driving situation on the basis of ambient conditions (temperature, etc.) and the current drivetrain situation (gear ratio, states). On this basis, the controller decides on the transition between active and passive state of the valuation function (see section "Valuation conditions of the valuation function" above). If the controller recognizes that the prerequisites for activating the evaluation function are not met, then step 401 carried out again. If, on the other hand, the controller recognizes that the prerequisites for activating the evaluation function are present (Y), then step 402 (PRECALC) where necessary precalculations are performed. For example, in step 402 the speed difference between electric machine and internal combustion engine calculated (slip). Further, at step 402 Determines whether the VKM is really at a standstill. After performing the pre-calculations in step 402 will with the steps 403 . 405 . 407 . 409 . 411 and 413 continued. These steps 403 . 405 . 407 . 409 . 411 and 413 implement a switch statement that recognizes various internal phases PHASE0, PHASE1, PHASE2, PHASE3, PHASE4 and PHASE5. Depending on the internal phase follows a corresponding status STAT0, STAT1, STAT2, STAT3, STAT4 or

STAT5 in the steps 404 . 406 . 408 . 410 . 412 and 414 , The switch statement detects the defined phases of a restart via the internal function status. After passing through the switch statement, the program returns via step 415 at the beginning 401 back.

In the case constellation of the step 403 (PHASE0) checks whether the boundary conditions are met. The evaluation function is waiting for a restart. In the case constellation of the step 405 (PHASE1) it is checked whether the restart starts with the filling of the clutch K0 (phase 1 of the 2 ). In the case constellation of the step 407 (PHASE2) it is checked whether the state of the separating coupling K0 changes from "open" to "transition". In the case constellation of the step 409 (PHASE3) it is checked whether the slip is built up. If this is the case (Y), it is done at step 410 (STAT4) an analysis and evaluation of the restart, as described in section "Monitoring the input speed" above.

In the case constellation of the step 411 (PHASE4), the K0 characteristic and the dynamics of the restart are plausibilized. In the case constellation of the step 413 (PHASE5) will be at step 414 (STAT4) performed an adjustment process depending on the evaluation (abnormality) of the restart.

The evaluation function is subdivided according to this embodiment into the following internal phases (not to be confused with those mentioned above in connection with FIG 2 described restart phases), each of which returns a functional status: Phase 0 Status 0 Function ready, waiting for restart request, pre-evaluations Phase 1 Status 1 Restart active and restart is in phase 1 Phase 2 Status 2 Disconnect switch from open to slipping, so restart in phase 2 Phase 3 Status 3 Observing and evaluating the EM speed Phase 4 Status 4 Synchronization of VKM finished. Plausibilisierung the VKM side moment balance Phase 5 Status 5 Nachlernstrategie: Decide whether to be re-learned

LIST OF REFERENCE NUMBERS

1

Internal combustion engine (VKM)

2

electric machine (EM)

3

transmission

4

differential

5

Disconnect coupling (K0)

6

High-voltage battery

7

power electronics

8th

Dual Mass Flywheel

ω. _EM

Speed gradient of the electric machine

ω _EM

Speed of the electric machine

ω _GEW

Speed of the active transmission input shaft

ω. _VKM

Speed gradient of the internal combustion engine

B _Abs

Evaluation variable regarding drive speed

B _degrees

Evaluation variable regarding speed gradient

B _Ges

total evaluation size

Factor _abs

Weighting factor speed

Factor _degrees

Weighting factor gradient

Δ _should

predefined setpoint

M _K0

K0 actual torque

M _EM

Moment provided by electrical machine

M _{VKM, loss}

Loss moment of the internal combustion engine

M _comp

Compensation torque EM

M _{driver's request}

requested torque (driver request torque)

μ

Kupplungsreibwert

_m

average friction radius (coupling)

z

Number of friction surfaces (clutch)

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

• WO 2011/064018 [0003]
• DE 102008004366 A1 [0004, 0005]
• DE 102011084844 A1 [0004]

Claims (10)

Method comprising evaluating at least one compensated restart of an internal combustion engine ( 1 ) by detecting the time behavior between the clutch torque (M _K0 ) of a separating clutch ( 5 ) and that of an electric machine ( 2 ) provided compensation torque (M _Komp ) by evaluating the drive speed (ω _EM ) and / or the speed gradient (ω. _EM ) of the electric machine ( 2 ) during the construction of the compensation torque (M _comp ) and / or the clutch torque (M _Kup ). Method according to Claim 1, in which the rotational speed ω _{EM of} the electric machine is observed and evaluated as an offset (Δ) relative to the rotational speed (ω _GEW ) of an active transmission input shaft. Method according to one of the preceding claims, further comprising adapting the time behavior between the clutch torque (M _K0 ) and the compensation torque (M _Komp ). The method of any one of the preceding claims, further comprising post-adapting the time constant of the compensated restart. Method according to Claim 4, in which counter-restarting is detected via a counter and is only post-adapted after a certain number of noticeable re-starts. Method according to one of the preceding claims, further comprising evaluating the driving situation on the basis of environmental conditions and / or a current drive train situation, wherein based on this evaluation, a transition between active and passive state of a restart test takes place. The method of any one of the preceding claims, further comprising detecting defined phases of a restart. A method according to any one of the preceding claims, further comprising calculating the transmitted clutch torque across the speed gradient (ω, _VKM ) of the internal combustion engine and plausibility check of the clutch characteristic. The method of any one of the preceding claims, further comprising evaluating the change in the driver's desired torque and other environmental conditions during a restart. Method according to one of the preceding claims, further comprising detecting overfilling of an electro-hydraulically actuated separating clutch during the filling phase at the Kisspoint pressure level by evaluating the rotational speed gradient (ω, _EM ) of the electric machine and the rotational speed (ω, _VKM ) of the internal combustion engine, and / or an evaluation and an adaptation of the applicative target specification of the filling pressure of the electro-hydraulic actuated separating clutch.

Images