METHOD AND DEVICE FOR CONTROLLING AN INTERNAL COMBUSTION ENGINE

Abstract

The invention relates to a method for stopping an internal combustion engine, wherein an amount of air which is supplied via an air metering device of the internal combustion engine, in particular a throttle flap (100), is reduced after a stopping order has been detected. According to the invention, the amount of air which is supplied via the air metering device of the internal combustion engine is again increased when the detected speed (n) of the internal combustion engine falls below a predefinable speed threshold value (ns), wherein an intake cylinder (ZYL2) to which the amount of air is supplied does not enter any working cycle after the amount of supplied air has been increased.

Description

BACKGROUND OF THE INVENTION

Particularly in the case of vehicles with start/stop technology, i.e. when the engine is frequently switched off and on again during normal driving operation, comfortable running down of the internal combustion engine and rapid restarting of the internal combustion engine is of great importance.

JP-2008298031 A describes a method in which the throttle valve of the internal combustion engine is closed during rundown in order to suppress vibration. By means of this measure, the air charge in the cylinders in the internal combustion engine is reduced, thus reducing the roughness of rundown since compression and decompression are minimized.

To restart the internal combustion engine, however, as much air as possible is required in the cylinders in which ignition takes place for the restart. There is therefore a conflict of aims between rapid engine starting (which requires a large quantity of air in the cylinder) and comfortable, i.e. low-vibration, engine rundown (which requires a small amount of air in the cylinder). This conflict of aims is resolved by means of the present invention.

Devices which modify the stroke profile particularly of the inlet valves of the internal combustion engine and thus adjust the air charge in the cylinders are common knowledge in the prior art. In particular, the fact that the stroke profile of the inlet valves can be configured as desired within wide limits by means of electrohydraulic actuators is known. Internal combustion engines with such electrohydraulic valve adjustment do not require a throttle valve. It is likewise known that the stroke profile, particularly of the inlet valves, can be varied by adjusting the camshaft. Devices of this kind and the throttle valve, with which the air charge in the cylinders can be modified, are also referred to below as air metering devices.

SUMMARY OF THE INVENTION

If a quantity of air supplied to the internal combustion engine is reduced by means of an air metering device and only increased again shortly before the internal combustion engine comes to a halt, “engine shake”, i.e. the generation of discernible vibration, can be avoided. This is achieved by initially reducing the quantity of air supplied to the internal combustion engine as the internal combustion engine runs down and increasing it again if a detected speed of the internal combustion engine has fallen below a speed threshold value.

An increased quantity of air is then supplied to an inlet cylinder which is in an intake stroke immediately after or during the increase in the quantity of air supplied, and it then has an increased air charge. If this inlet cylinder then goes into a compression stroke, the increased air charge acts as a gas spring, which exerts a high restoring torque on a crankshaft via the inlet cylinder ZYL2. Conversely, the respective air charge in the cylinders which go into a downward movement exerts a torque on the crankshaft acting in the direction of the forward rotation of the crankshaft. However, since these cylinders going into a downward movement have a small air charge, the overall torque acting on the crankshaft is a restoring torque.

If the speed threshold value is suitably chosen, it is possible to ensure that the inlet cylinder no longer goes into a power stroke after the increase in the quantity of air metered in. This has the advantage that compression of the increased air charge is avoided, preventing unwanted vibration.

It is particularly advantageous if the speed threshold value is selected in such a way that the inlet cylinder just fails to go into the power stroke after the increase in the quantity of air metered in. If the speed threshold value is selected in such a way and if the speed of the internal combustion engine is higher than the speed threshold value when a request for restarting is detected, it is possible to implement a method for particularly rapid restarting of the internal combustion engine.

In order to reliably select precisely the speed threshold value which ensures that the inlet cylinder just fails to go into the power stroke after the increase in the quantity of air metered in, the invention proposes an adaptation method. For this purpose, it is necessary to define suitable criteria, according to which the speed threshold value is reduced or increased.

Reducing the speed threshold value if the inlet cylinder still passes through a top dead center position after the increase in the quantity of air metered in and before the internal combustion engine comes to a halt is a particularly simple way of ensuring that vibration due to impermissible passage through a top dead center position at a high air charge is suppressed during the subsequent operation of the internal combustion engine.

Increasing the speed threshold value if the inlet cylinder no longer goes into a compression stroke after the increase in the amount of air metered in is a particularly simple way of ensuring that the inlet cylinder exhibits an oscillatory behavior when stopping during the subsequent operation of the internal combustion engine.

Modifying the speed threshold value in accordance with a reverse oscillation angle is a particularly simple way of ensuring that the inlet cylinder exhibits a defined oscillatory behavior in the future operation of the internal combustion engine.

Increasing the speed threshold value if the reverse oscillation angle is less than a specifiable minimum reverse oscillation angle ensures that the inlet cylinder just fails to reach the top dead center position with a particularly high degree of reliability.

If the speed threshold value is increased to a specifiable initial threshold value, the adaptation method according to the invention has defined entry points and is therefore particularly robust.

If the selected magnitude of the initial threshold value is such that the inlet cylinder reliably passes through the top dead center position, this ensures that the speed threshold value ns is always adapted starting from values that are too high, making the adaptation method particularly simple.

The dead center positions are the simplest points at which to monitor the speed of the internal combustion engine. If the system determines, at one dead center position, that the speed has fallen below the speed threshold, the inlet cylinder is just going into the inlet stroke. If the quantity of air metered in by the air metering device is increased while the outlet valve of the inlet cylinder is still open, an increased quantity of air is pumped into an exhaust pipe from an intake pipe. This leads to disadvantageous noise generation. If, on the other hand, the quantity of air metered in by the air metering device is increased too late during the inlet stroke of the inlet cylinder, there is a high pressure drop between the intake pipe and the cylinder. In this case, the inflow of air leads to considerable unwanted noise generation. To minimize this noise generation, it is advantageous if the quantity of air metered in by the air metering device is increased immediately after the end of valve overlap in the inlet cylinder, i.e. immediately after the closure of the outlet valve.

Since the internal combustion engine is halted, fuel injection is switched off. For rapid restarting of the internal combustion engine, this is disadvantageous since the cylinders do not contain an ignitable mixture. Since, in the method according to the invention, air is passed into the inlet cylinder from the intake pipe, it is possible to ensure, given appropriate injection before the end of the inlet stroke, that there is an ignitable fuel/air mixture in the inlet cylinder. Since the inlet cylinder comes to rest in the vicinity of a bottom dead center position or in the compression stroke, this is very advantageous for a rapid restart since a starter has to carry out a rotation of the crankshaft of just 180° before ignition can take place in the inlet cylinder.

If the fuel is injected before or immediately after the inlet cylinder goes into the inlet stroke, this is particularly advantageous for mixture formation. In the case of intake pipe injection, the amount of fuel metered in can be particularly finely metered and, in the case of direct injection, early injection of fuel is advantageous for the turbulent mixing of air and fuel.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are explained in greater detail below with reference to the attached drawings, in which:

FIG. 1 shows an illustration of a cylinder of an internal combustion engine,

FIG. 2 shows schematically the profile of a number of characteristic quantities of the internal combustion engine as the internal combustion engine is stopped,

FIG. 3 shows the sequence of the method according to the invention for stopping the internal combustion engine,

FIG. 4 shows a speed profile during the stopping and restarting of the internal combustion engine,

FIG. 5 shows a detailed view of the speed profile during the stopping and restarting of the internal combustion engine,

FIG. 6 shows the sequence of the method according to the invention during the restarting of the internal combustion engine,

FIG. 7 shows schematically a final oscillatory motion of the internal combustion engine at different speed threshold values, and

FIG. 8 shows the sequence of the method according to the invention for determining the speed threshold value.

DETAILED DESCRIPTION

FIG. 1 shows a cylinder 10 of an internal combustion engine having a combustion chamber 20, a piston 30, which is connected by a connecting rod 40 to a crankshaft 50. The piston 30 performs an up and down motion in a known manner. The reversal points of the motion are referred to as dead center positions. The transition from an upward motion to a downward motion is referred to as the top dead center position, while the transition from a downward motion to an upward motion is referred to as the bottom dead center position. An angular position of the crankshaft 50, referred to as a crank angle, is conventionally defined relative to the top dead center position. A crankshaft sensor 220 detects the angular position of the crankshaft 50.

Air to be combusted is sucked into the combustion chamber 20 via an intake pipe 80 in a known manner during a downward motion of the piston 30. This is referred to as the intake stroke or inlet stroke. The combusted air is forced out of the combustion chamber 20 via an exhaust pipe 90 during an upward motion of the piston 30. This is usually referred to as the exhaust stroke. The quantity of air sucked in via the intake pipe 80 is set by means of an air metering device, in the illustrative embodiment a throttle valve 100, the position of which is determined by a control device 70.

Via an intake pipe injection valve 150, which is arranged in the intake pipe 80, fuel is injected into the air sucked out of the intake pipe 80, and a fuel/air mixture is produced in the combustion chamber 20. The quantity of fuel injected through the intake pipe injection valve 150 is determined by the control device 70, generally by means of the duration and/or level of an activation signal. A spark plug 120 ignites the fuel/air mixture.

An inlet valve 160 at the inlet from the intake pipe 80 to the combustion chamber 20 is driven via cams 180 by a camshaft 190. An outlet valve 170 at the inlet from the exhaust pipe 90 to the combustion chamber 20 is likewise driven via cams 182 by the camshaft 190. The camshaft 190 is coupled to the crankshaft 50. The camshaft 190 generally performs one revolution for every two revolutions of the crankshaft 50. The camshaft 190 is designed in such a way that the outlet valve 170 opens in the exhaust stroke and closes in the vicinity of the top dead center position. The inlet valve 160 opens in the vicinity of the top dead center position and closes in the inlet stroke. A phase in which the outlet valve 170 and the inlet valve in one system are opened simultaneously is referred to as valve overlap. Such valve overlap is used for internal exhaust gas recirculation, for example. The camshaft 190 can be designed, in particular, for activation by the control device 70, making it possible to set different stroke profiles for the inlet valve 160 and the outlet valve 170 in accordance with the operating parameters of the internal combustion engine. However, it is also possible for the inlet valve 160 and the outlet valve 170 not to be moved up and down by means of the camshaft 190 but by means of electrohydraulic valve actuators. In this case, the camshaft 190 and the cams 180 and 182 can be omitted. There is likewise no need for the throttle valve 100 with such electrohydraulic valve actuators.

A starter 200 can be connected mechanically to the crankshaft 50 by a mechanical coupling 210. The production of the mechanical connection between the starter 200 and the crankshaft 50 is also referred to as meshing. Release of the mechanical connection between the starter 200 and the crankshaft 50 is also referred to as disengagement. Meshing is possible only if the speed of the internal combustion engine is below a speed threshold value dependent on the internal combustion engine and the starter.

FIG. 2 shows the behavior of the internal combustion engine as the internal combustion engine is stopped. FIG. 2a shows the sequence of the various strokes of a first cylinder ZYL1 and of a second cylinder ZYL2, plotted against the angle of the crankshaft KW. A first dead center position T1, a second dead center position T2, a third dead center position T3, a fourth dead center position T4 and a fifth dead center position T5 of the internal combustion engine are plotted. Between these dead center positions, the first cylinder ZYL1 runs through the exhaust stroke, the inlet stroke, a compression stroke and a power stroke in a known manner. In the illustrative embodiment of an internal combustion engine having four cylinders, the strokes of the second cylinder ZYL2 are offset by 720°/4=180°. Based on the first cylinder ZYL1, the first dead center position T1, the third dead center position T3 and the fifth dead center position T5 are bottom dead center positions, while the second dead center position T2 and the fourth dead center position T4 are top dead center positions. Based on the second cylinder ZYL2, the first dead center position T1, the third dead center position T3 and the fifth dead center position T5 are top dead center positions, while the second dead center position T2 and the fourth dead center position T4 are bottom dead center positions.

FIG. 2b shows the profile of a speed n of the internal combustion engine against time t in parallel with the strokes illustrated in FIG. 2a. The speed n is defined as the time derivative of the crank angle KW, for example. The first dead center position T1 corresponds to a first time t1, the second dead center position T2 corresponds to a second time t2, the third dead center position T3 corresponds to a third time t3, and the fourth dead center position T4 corresponds to a fourth time t4. Between each two successive times, e.g. between the first time t1 and the second time t2, the speed initially rises briefly, and then falls monotonically. The brief rise in speed is due to the compression of the air charge in the cylinders. A cylinder running through a top dead center position compresses the air charge therein to the maximum extent, and therefore compression energy is stored therein. Part of this compression energy is converted into rotational energy as the internal combustion engine continues to rotate.

FIG. 2c shows the time profile of an activation signal DK of the throttle valve 100 in parallel with FIG. 2a and FIG. 2b. As is known from the prior art, the throttle valve 100 is initially closed as the internal combustion engine is stopped, this corresponding to a first activation signal DK1. If, as illustrated in FIG. 2b, the speed n of the internal combustion engine falls below a speed threshold value ns, e.g. 300 rpm, then, according to the invention, the throttle valve 100 is opened at an opening time tauf, corresponding to a second activation signal DK2. Here, the opening time tauf is selected in such a way that it occurs shortly after the third dead center position T3, which is the next dead center position after the speed n of the internal combustion engine falls below the speed threshold value ns. At the third dead center position T3, the second cylinder ZYL2 goes into the inlet stroke. In what follows, therefore, it is also referred to as inlet cylinder ZYL2. In the illustrative embodiment, the opening time tauf coincides with the end of valve overlap in the inlet cylinder, i.e. with the time at which the outlet valve 170 of the inlet cylinder ZYL2 closes. Based on the top dead center position of the inlet cylinder ZYL2, the opening time tauf corresponds to an opening crank angle KWauf. To determine the time at which the speed n of the internal combustion engine has fallen below the speed threshold value ns, the speed n of the internal combustion engine can either be monitored continuously. Since the rise in the speed n of the internal combustion engine is small after the dead center positions, and the opening time tauf is supposed to be shortly after a dead center position, however, it is also possible to check at each dead center position of the internal combustion engine whether the speed n of the internal combustion engine has fallen below the speed threshold ns. In the illustrative embodiment illustrated in FIG. 2b, the fact that the speed n of the internal combustion engine has not yet fallen below the speed threshold ns is detected at the first time t1 and the second time t2. At the third time t3, the system detects for the first time that the speed n of the internal combustion engine has fallen below the speed threshold ns, and the throttle valve 100 opens.

The opening of the throttle valve 100 then allows a large quantity of air to flow into the inlet cylinder in the inlet stroke. If the inlet cylinder ZYL2 goes into the compression stroke after the fourth time t4, the compression work to be performed on the air charge, which is greatly increased relative to the other cylinders, exceeds the compression energy released in the expanding cylinders, and the speed n of the internal combustion engine falls rapidly until it falls to zero at a reverse oscillation time tosc. The rotary motion of the crankshaft 50 is now reversed, and the speed n of the internal combustion engine becomes negative. The reverse oscillation time tosc corresponds to a reverse oscillation angle RPW of the crankshaft 50 which is indicated in FIG. 2a. At a stop time tstopp, the internal combustion engine comes to a halt. It should be noted that the time axis is depicted in a nonlinear manner. In accordance with the drop in the speed n of the internal combustion engine, the time interval between the third time t3 and the fourth time t4 is longer than the time interval between the second time t2 and the third time t3, which in turn is longer than the time interval between the first time t1 and the second time t2. The fifth dead center position T5 of the internal combustion engine is not reached. In the time interval between the reverse oscillation time tosc and the stop time tstopp, the crankshaft 50 performs an oscillatory motion, during which the second cylinder ZYL2 oscillates in the compression stroke and the inlet stroke thereof, while the first cylinder ZYL1 oscillates in a corresponding manner in the power stroke and the compression stroke thereof.

FIG. 3 shows the sequence of the method, which corresponds to the method illustrated in FIG. 2. With the internal combustion engine running, it is determined in a stop detection step 1000 that the intention is to switch off the internal combustion engine. This is followed by step 1010, in which injection and ignition are switched off. The internal combustion engine is thus in the rundown mode. There then follows step 1020, in which the throttle valve is closed. In the case of internal combustion engines with camshaft adjustment, a switchover to a smaller cam can take place in step 1020 as an alternative, thus reducing the air charge in the cylinders. In the case of internal combustion engines with electrohydraulic valve adjustment, the valves of the internal combustion engine can be closed in step 1020. There follows step 1030, in which the system checks whether the speed n of the internal combustion engine has fallen below the speed threshold value ns. If this is the case, step 1040 follows. If this is not the case, step 1030 is repeated until the speed n of the internal combustion engine has fallen below the speed threshold value ns. In step 1040, the throttle valve 100 is opened at opening time tauf. In the case of internal combustion engines with camshaft adjustment, it is possible instead for a switch to be made to a larger cam in step 1040, for example, resulting in an increase in the air charge in the inlet cylinder ZYL2. In the case of internal combustion engines with electrohydraulic valve adjustment, the inlet valve 160 of the inlet cylinder ZYL2 can be activated in such a way in step 1040 that it is open during the inlet stroke of the inlet cylinder ZYL2, thus increasing the air charge in the inlet cylinder ZYL2. There follows step 1060. In the optional step 1060, fuel is injected via the intake pipe injection valve 150 into the intake pipe 80 of the internal combustion engine. This injection of fuel is performed in such a way that a fuel/air mixture is sucked into the inlet cylinder ZYL2 in the inlet stroke. In step 1100, the method according to the invention ends. As illustrated in FIG. 2b, the internal combustion engine oscillates into a stationary position, in which the inlet cylinder ZYL2 comes to rest in the inlet stroke or in the compression stroke. Injection of fuel in step 1060 is advantageous for rapid restarting of the internal combustion engine when it is an internal combustion engine with intake pipe injection.

FIG. 4 shows the time profile of the speed n of the internal combustion engine when stopping and restarting. The speed n of the internal combustion engine falls during a rundown phase T_Auslauf in the manner illustrated in FIG. 2b, and finally the sign changes when the rotary motion of the internal combustion engine is reversed at the reverse oscillation time tosc illustrated in FIG. 2b. This is illustrated in FIG. 4 as the end of the rundown phase T_Auslauf and the beginning of an oscillation phase T_Pendel. While the rundown phase T_Auslauf is still ongoing, the system determines at a starting request time tstart that the internal combustion engine is to be restarted because, for example, the system has detected that a driver has pressed a gas pedal. A determined start request of this kind before the stop time tstopp, is also referred to as a “change of mind”. In the oscillation phase T_Pendel, the profile of the speed n of the internal combustion engine undergoes a resulting variation until it falls to a constant zero at the stop time tstopp illustrated in FIG. 2b and remains there. In FIG. 4, the stop time tstopp marks the end of the oscillation phase T_Pendel.

In the prior art method for starting the internal combustion engine, the oscillation phase T_Pendel is followed by detection of the fact that the internal combustion engine is stationary, the starter 200 is meshed, and the starter is activated. After an activation dead time T_tot of the starter 200 of, for example, 50 ms, which is not illustrated in FIG. 4, the starter 200 begins a rotary motion at a time tSdT and thus imparts motion to the crankshaft 50 once again. In the method according to the invention, in contrast, a first meshing time tein1 and, if appropriate, a second meshing time tein2 is determined. The first meshing time tein1 and the second meshing time tein2 are characterized in that the speed n of the internal combustion engine is sufficiently low for the starter 200 to be meshed. The first meshing time tein1 and the second meshing time tein2 are determined by the control device 70. If the time interval between the starting request time tstart and the first meshing time tein1 is longer than the activation dead time T_tot, the starter 200 is meshed and activated in such a way that it begins a rotary motion at the first meshing time tein1. If the first meshing time tein1 is too close in time to the starting request time tstart, the starter 200 is meshed and activated in such a way that it begins a rotary motion at the second meshing time tein2.

FIG. 5 illustrates in detail the selection of the first meshing time tein1 and the second meshing time tein2. As described, the speed n of the internal combustion engine falls rapidly to zero after the opening time tauf, and the internal combustion engine begins a reverse motion at reverse oscillation time t_osc. The first meshing time tein1 is determined by means of characteristic maps or by means of models stored in the control device 70, for example, after the opening of the throttle valve 100 and corresponds to the estimated reverse oscillation time tosc. It is, of course, also possible for different times at which the speed n of the internal combustion engine passes through zero to be predicted and selected as the first meshing time tein1 instead of the reverse oscillation time tosc.

In addition to the passage of the speed n of the internal combustion engine through zero, a second meshing time tein2 can be selected, from which time onwards it is ensured that the speed n of the internal combustion engine will no longer leave a speed range in which meshing of the starter 200 is possible. This speed range is given, for example, by a positive threshold nplus, e.g. 70 rpm, up to which the starter 200 can be meshed during a forward rotation of the internal combustion engine, and by a negative threshold nminus, e.g. 30 rpm, up to which the starter 200 can be meshed during a reverse rotation of the internal combustion engine. Using characteristic maps, for example, the control device 70 calculates that the kinetic energy of the internal combustion engine has fallen from the second meshing time tein2 to such an extent that the speed range [nminus, nplus] will no longer be exceeded. At the second meshing time tein2 or at any time after the second meshing time tein2, the starter 200 can be meshed and made to perform a rotary motion.

FIG. 6 shows the sequence of the method according to the invention for restarting the internal combustion engine. Step 2000 coincides with step 1000 illustrated in FIG. 3. In this step, a request to stop the internal combustion engine is determined. There follows step 2005. In step 2005, the throttle valve is closed, or other measures, e.g. adjustment of the cams 180, 182 or appropriate electrohydraulic activation of the valves 160 and 170, are taken in order to reduce the air charge in the cylinders. There follows step 2010.

In step 2010, the system determines whether a start request for starting the internal combustion engine is determined while the internal combustion engine is still running down, i.e. during the rundown phase T_Auslauf illustrated in FIG. 4. If this is the case, step 2020 follows. If this is not the case, step 2090 follows. In step 2020, the system checks whether the speed n of the internal combustion engine is above the speed threshold value ns (if appropriate by a minimum amount, e.g. 10 revolutions per minute). These checks can take place continuously or in synchronism with the crankshaft, in particular at each dead center position of the internal combustion engine. If the speed n of the internal combustion engine is above the speed threshold value ns, step 2030 follows and otherwise step 2070 follows.

In step 2030, the throttle valve is opened, or other measures, e.g. adjustment of the cams 180, 182 or appropriate electrohydraulic activation of the valves 160 and 170, are taken in order to increase the air charge in the cylinder which is the next to be in the inlet stroke. Via the intake pipe injection valve 50, fuel is injected into the intake pipe 80. There follows step 2040, in which the inlet cylinder ZYL2 is determined, i.e. the cylinder in which the air charge will be the next to show a significant increase in the inlet stroke. The inlet cylinder ZYL2 goes into the inlet stroke and sucks in the fuel/air mixture in the intake pipe 80. The inlet cylinder ZYL2 then makes a transition to the compression stroke. The speed n is higher than the speed threshold value ns. The speed threshold value ns is selected in such a way that the inlet cylinder ZYL2 just fails to pass through a top dead center position. At the speed n of the internal combustion engine, it is therefore ensured that the inlet cylinder ZYL2 passes through a top dead center position once again and makes a transition to the power stroke. There follows step 2050. In step 2050, the fuel/air mixture in the inlet cylinder ZYL2 is ignited, accelerating the rotation of the crankshaft 50, and step 2060 follows. In step 2060, further measures are carried out in order to bring about starting of the internal combustion engine, in particular a fuel/air mixture being ignited in a corresponding manner in the other cylinders of the internal combustion engine. With the starting of the internal combustion engine, the method according to the invention ends.

In step 2070, fuel is injected into the intake pipe 80 via the intake pipe injection valve 150. There follows step 2100.

In step 2090, the system checks, in a manner corresponding to step 1030 illustrated in FIG. 3, whether the speed n of the internal combustion engine has fallen below the speed threshold value ns. If this is not the case, the program branches back to step 2010. If this is the case, step 2100 follows.

Step 2100 corresponds to step 1040 in FIG. 3. The throttle valve is opened or some other air metering device, e.g. a camshaft adjustment system or an electrohydraulic valve timing system, is activated in such a way that the quantity of air supplied is increased. There follows step 2110.

In step 2110, the system determines whether there is a request for starting the internal combustion engine. If this is the case, step 2120 follows. If this is not the case, step 2110 is repeated until there is a request for starting the internal combustion engine. In step 2120, the system checks whether the internal combustion engine is stationary. This corresponds to the time period illustrated in FIG. 4 following the end of the oscillation phase T_Phase. If this is the case, step 2060 follows, in which conventional measures for starting the internal combustion engine are carried out. As illustrated in FIG. 4, the internal combustion engine is started at a time tSdT.

If the internal combustion engine is not stationary in step 2120, step 2150 follows. In step 2150, the first meshing time tein1 is predicted. This prediction is performed by means of a characteristic map, for example. Using the speed n which was determined during a previous passage through the top dead center position of the inlet cylinder ZYL2 (at the fourth time t4 in the illustrative embodiment), the kinetic energy of the internal combustion engine can be determined and, from the second position DK2 of the air metering device, the air charge in the inlet cylinder ZYL2 and hence the strength of the gas spring compressed by the inlet cylinder ZYL2 in the compression stroke can be estimated. From this, it is possible to estimate the reverse oscillation time tosc, which is predicted as the first meshing time tein1. There follows step 2160, in which the system checks whether the time difference between the first meshing time tein1 and the present time is greater than the activation dead time T_tot of the starter 200. If this is the case, step 2170 follows. If this is not the case, step 2180 follows.

In step 2180, the second meshing time tein2 is determined. As explained in FIG. 5, the second meshing time tein2 is selected in such way that the speed n of the internal combustion engine from the second meshing time tein2 onwards remains in the speed interval between the negative threshold nminus and the positive threshold nplus. In the following step 2190, the starter 200 is meshed and starting is carried out from the second meshing time tein2. There follows step 2060, in which the further measures for starting the internal combustion engine are carried out. As an alternative, it is also possible, in step 2180, to determine a meshing interval, during which the speed n remains between the negative threshold nminus and the positive threshold nplus. In this case, the starter 200 is meshed and starting carried out in the meshing interval in step 2190.

Instead of an intake pipe injection valve 150, it is also conceivable for injection valves of the internal combustion engine to be arranged in the combustion chamber, i.e. to be configured as a direct injection valve. In this case, injection of fuel into the intake pipe immediately after the opening of the throttle valve can be omitted. The only factor of importance is that fuel should be injected in a suitable manner into the inlet cylinder ZYL2 before it is ignited upon restarting.

FIG. 7 illustrates the selection of the speed threshold value ns. FIG. 7a illustrates the oscillatory behavior of the inlet cylinder ZYL2 when the speed threshold value ns is correctly selected. At the opening crank angle KWauf, the inlet cylinder ZYL2 is in forward motion, passes through the bottom dead center position UT corresponding to the fourth dead center position T4 and reverses its direction of rotation at the reverse oscillation angle RPW. The further oscillatory motion of the inlet cylinder ZYL2 up to the stationary condition is shown only indicatively in FIG. 7a.

FIG. 7b illustrates the oscillatory behavior of the inlet cylinder ZYL2 if the speed threshold value ns selected is too high. A speed threshold value ns which is too high means that the kinetic energy of the internal combustion engine is too high when the throttle valve 100 is opened, i.e. at the opening crank angle KWauf. This leads to the inlet cylinder ZYL2 passing through the bottom dead center position UT corresponding to the fourth dead center position T4 and then also the top dead center position OT corresponding to the fifth dead center position T5. This leads to unwanted vibration in the drive train, and is felt to be uncomfortable by the driver.

FIG. 7c illustrates the oscillatory behavior of the inlet cylinder ZYL2 if the speed threshold value ns selected is too low. A speed threshold value ns which is too low means that the kinetic energy of the internal combustion engine is too low when the throttle valve 100 is opened, i.e. at the opening crank angle KWauf. The inlet cylinder ZYL2 passes through the bottom dead center position UT corresponding to the fourth dead center position, but has a relatively large reverse oscillation angle RPW. If, in step 3020, it is determined that the speed n of the internal combustion engine is higher than the speed threshold value ns, it is no longer safe to assume that the inlet cylinder ZYL2 will rotate beyond the top dead center position OT and hence that it will be possible to start the internal combustion engine quickly.

The selection of the speed threshold value ns is therefore of central importance for the functioning of the method according to the invention but, on the other hand, it is very difficult since it depends on variables which change during the life of the internal combustion engine, e.g. the friction coefficient of the engine oil used.

FIG. 8 describes an adaptation method, by means of which an initially specified speed threshold value ns can be adapted in order to compensate for errors in the initialization or changes in the properties of the internal combustion engine. In step 3000, it is determined that there is a stop request to the internal combustion engine, and measures for starting the internal combustion engine are initiated. In step 3010, the system checks, in a manner corresponding to step 1030, whether the speed n of the internal combustion engine has fallen below the speed threshold ns. If this is the case, step 3020 follows, in which the throttle valve is opened in a manner corresponding to step 1040. There follows step 3030, in which the system checks whether the inlet cylinder ZYL2 has already passed through the bottom dead center position UT. If this is not the case, step 3040 follows. If it is the case, step 3060 follows.

Step 3040 takes account of the case where the speed threshold value ns selected is so low that the internal combustion engine comes to a halt even before the inlet cylinder ZYL2 passes through the bottom dead center position UT. For this purpose, the system checks in step 3040 whether the internal combustion engine is stationary. If this is not the case, the program branches back to step 3030. If the internal combustion engine is stationary, step 3050 follows. In step 3050, the speed threshold value ns is increased. There follows step 3100, with which the method ends.

In step 3060, the rotary motion of the internal combustion engine is monitored. If the internal combustion engine turns the inlet cylinder ZYL2 further beyond the top dead center position OT, step 3070 follows. If the top dead center position OT is not reached, step 3080 follows. In step 3070, the behavior is as illustrated in FIG. 7b, and the speed threshold value ns is reduced. There follows step 3100, with which the method ends.

In step 3080, the reverse oscillation angle RPW is determined by means of the crankshaft sensor 220, for example. There follows step 3090. In step 3090, the system checks whether the reverse oscillation angle RPW is smaller than a minimum reverse oscillation angle RPWS, which is 10° for example. If the reverse oscillation angle RPW is smaller than the minimum reverse oscillation angle RPWS, the correct behavior shown in FIG. 7a is present, and step 3100 follows, with which the method ends. If the reverse oscillation angle RPW is larger than the minimum reverse oscillation angle RPWS, the behavior illustrated in FIG. 7c is present, and step 3050 follows, in which the speed threshold value ns is increased.

The increase in the speed threshold value ns in step 3050 can either take place incrementally or the speed threshold value ns is increased to an initial threshold value nsi, at which it is ensured that the internal combustion engine exhibits the behavior illustrated in FIG. 7b, i.e. that the speed threshold value ns selected is then initially too high. The initial threshold value nsi can be designed as an applicable threshold value, for example. It is selected in such a way that, within the scope of the operating parameters that are possible during the operation of the internal combustion engine, e.g. variations in the leakage of the air charge, differences in the engine oil or individual differences in the scatter of the frictional effect of the internal combustion engine, the internal combustion engine exhibits the behavior illustrated in FIG. 7b, i.e. that the inlet cylinder ZYL2 goes into the power stroke.

As an option, it is also possible for the adaptation of the speed threshold value ns to be carried out when restarting of the internal combustion engine has not taken place correctly: the speed threshold value ns is increased if the system has decided in step 2020 that the determined speed n of the internal combustion engine is higher than the speed threshold value ns and if, after steps 2030, 2040 and 2050 are carried out, it is ascertained in step 2060 that the inlet cylinder ZYL2 (ZYL2) has not gone into the power stroke.

Claims

1. A method for stopping an internal combustion engine, in which a quantity of air supplied to the internal combustion engine via an air metering device is reduced after a stop request has been detected, characterized in that the quantity of air supplied to the internal combustion engine via the air metering device is increased again if a detected speed (n) of the internal combustion engine falls below a specifiable speed threshold value (ns), wherein an inlet cylinder (ZYL2), to which the quantity of air is supplied, no longer goes into a power stroke after the quantity of air supplied is increased.

2. The method as claimed in claim 2, characterized in that the speed threshold value (ns) is reduced if the inlet cylinder (ZYL2) goes into the power stroke after the quantity of air metered in is increased and before the internal combustion engine comes to a halt.

3. The method as claimed in claim 1, characterized in that the speed threshold value is increased if the inlet cylinder (ZYL2) no longer goes into a compression stroke after the quantity of air metered in is increased.

4. The method as claimed in claim 1, characterized in that the specifiable speed threshold value is modified in accordance with a reverse oscillation angle (RPW).

5. The method as claimed in claim 4, characterized in that the speed threshold value is increased if the reverse oscillation angle (RPW) is greater than a specifiable minimum reverse oscillation angle (RPWS).

6. The method as claimed in claim 5, characterized in that the specifiable speed threshold value (ns) is increased to a specifiable initial threshold value (nsi).

7. The method as claimed in claim 7, characterized in that the selected magnitude of the initial threshold value (nsi) is such that the inlet cylinder (ZYL2) passes through the top dead center position.

8. The method as claimed in claim 1, characterized in that the quantity of air metered in by the air metering device is increased immediately after the closure of an outlet valve (160) of the inlet cylinder (ZYL2).

9. The method as claimed in claim 1, characterized in that fuel is injected in such a way that an ignitable fuel/air mixture is present in the inlet cylinder (ZYL2) when it leaves the inlet stroke.

10. The method as claimed in claim 1, characterized in that fuel is injected before the inlet cylinder (ZYL2) goes into the inlet stroke.

11. A computer program, characterized in that it is programmed for use in a method as claimed in claim 1.

12. An electric storage medium for an open-loop and closed-loop control device for an internal combustion engine, characterized in that a computer program for use in a method as claimed in claim 1 is stored on said medium.

13. An open-loop and closed-loop control device for an internal combustion engine, characterized in that it is programmed for use in a method as claimed in claim 1.

14. The method as claimed in claim 1, wherein the air metering device is a throttle valve (100).

15. The method as claimed in claim 1, characterized in that fuel is injected immediately after the inlet cylinder (ZYL2) goes into the inlet stroke.

16. An electric storage medium for an open-loop control device for an internal combustion engine, characterized in that a computer program for use in a method as claimed in claim 1 is stored on said medium.

17. An electric storage medium for a closed-loop control device for an internal combustion engine, characterized in that a computer program for use in a method as claimed in claim 1 is stored on said medium.

18. An open-loop control device for an internal combustion engine, characterized in that it is programmed for use in a method as claimed in claim 1.

19. A closed-loop control device for an internal combustion engine, characterized in that it is programmed for use in a method as claimed in claim 1.