DE19902203A1 - Determining ignition time point of internal combustion engine involves determining ignition time point from characteristic field using future load signal predicted from detected load signal - Google Patents

Abstract

The method involves generating a characteristic field for the ignition time point depending at least on a load signal indicating the relative air filling of a cylinder of the engine, detecting the load signal at a time before the determined time point, predicting a future load signal at a later time before the determined time point; and determining the ignition time point from the characteristic field using the predicted future load signal. An Independent claim is also included for an arrangement for determining the ignition time point of an internal combustion engine.

Description

STATE OF THE ART

The present invention relates to a method and a Device for determining the ignition timing of a burner engine.

Although applicable to any internal combustion engine the present invention and the underlying Problems with an internal combustion engine Motor vehicle explained.

In the sense of optimally controlling a burner The engine should have the relative air filling of the cylinders an internal combustion engine to be known as precisely as possible, because with a force precisely matched to this air filling amount of substance can be metered and thus the desired Torque with low pollutant emissions and low Fuel consumption can be achieved. An optimal one Fuel metering is complicated by the fact that Time from which the actual air filling of a Zylin which is known to the internal combustion engine, the fuel measurement for this cylinder has already been completed.  

In other words, for the fuel metering in generally used outdated values of air filling. If the air filling does not change from intake stroke to intake stroke or changes only slightly, can also become obsolete with such values for air filling are optimal or almost optimal fuel metering can be achieved. In operation conditions in which the air filling changes very strongly, however, it is cheaper to adjust the fuel metering to the to coordinate the expected air filling.

For this purpose, a method is specified in DE 44 01 828 A1 been that at the time of calculating the attributable Fuel quantity an accurate forecast of the air Fills the cylinder in which the fuel volume ge is injected.

According to the teaching of DE 44 01 828 A1, a future Load signal determined that the expected relative air filling represents. The future load signal will be off a current main load signal, a current auxiliary load signal that leads the current main load signal, and a crank angle interval. The crank angle The interval depends on the time units or course fuel angle storage expressed in fuel angle units, the duration of the fuel injection and the calculation time can be specified. The inclusion of the crank angle interval has the advantage that the determination of the future load si be carried out at the latest possible time  can and thus a high accuracy is achieved.

It is useful that the future load signal with a Low pass filter is determined, its filter constant can be specified depending on the load. The filter constant is at reading increasing load from a first characteristic and with falling load from a second characteristic. Thereby is a particularly computing time-saving predetermination of the Air filling possible.

The auxiliary load signal is derived from the opening angle of the Dros selklappe, the speed of the internal combustion engine and one if necessary through a bypass channel to the throttle valve flowing air volume determined and depending on the tempe air intake and barometric altitude rigged.

With small opening angles of the throttle valve, this can Auxiliary load signal also from the with an air mass meter air mass can be determined, which usually leads to egg leads to higher accuracy in this operating range.

The main load signal can e.g. B. from the measured intake manifold pressure and the speed from which with an air mass meter detected air mass or by filtering the auxiliary load gnals can be determined.

The method can be used both in non-stationary operation can also be used in stationary operation because the  Determination of the future load signal on the main load signal matched auxiliary load signal is used. The Ab required for the adjustment of the auxiliary load signal equivalence is achieved by integrating the difference between the main load signal and the one with the adjustment value filtered auxiliary load signal determined. The filtered Auxiliary load signal is corrected by filtering the Auxiliary load signal generated.

In this known method, the future Lastsi only for the determination of the injected Amount of fuel used.

The problem underlying the present invention is that in dynamic processes in the form of changes in the load at the time of calculation t _{0 of} the basic ignition angle ZWGRU z. B. air load that can be determined from the load signal tL sometimes deviates by up to 50% from the value actually present at the ignition time t _z .

This deviation is caused, among other things, by the dead time of approx. 50 ms between the time t _{0 of} the calculation of the air filling and the actual filling process, as illustrated in FIG. 3 (upper curve) by the load signal (tL) curve. Because of this deviation, which is referred to as an update error, corresponding dynamic corrections of filling-dependent variables are not only necessary at the injection level, but also at the ignition level. Even averaging over the time interval [t _-1 , t ₀ ], which would result in an averaged load signal tLM, would not bring any improvement here.

Corrections to the ignition parameters (ignition angle, closing angles), however, are currently usually excluded Lich with empirical, partly adaptive and complex Dyna micro-corrections at the manipulated variable level.

In Fig. 5 the curve with the open squares shows the knock limit of the ignition angle w _z as a function of the load signal tL at constant speed n. The knock limit can vary depending on further parameters, such as. B. intake air temperature, fuel / air ratio late or early, the basic position for the optimal ignition angle, which is shown in the curve with the closed squares in Fig. 5, is retained. Optimally means a coordination of torque, consumption, etc.

With increasing load, the knock limit shifts at a constant speed n to later (smaller) firing angles, as illustrated in FIG. 5. This tendency is also reflected in the basic ignition angle ZWGRU, which is stored, for example, as a function of the load signal tL and the speed n in a characteristic field KF, as illustrated in FIG. 6.

If the load signal is too small under load dynamics used to read the map KF, so in the result an early ignition angle ZWGRU output. In the fol If it were dynamic, it would increase to an undesirable level  Knocking frequency come.

So the usual solution is in knocking dynamics tries to output an adapted dynamic lead (= Ignition angle retard for the duration of the dynamic) this difference between this too early ZWGRU and the ei Occasionally necessary, correspond to the actual load signal equalize basic ignition angle ZWGRU.

Has been found to be disadvantageous in the above known approach highlighted the fact that the adaptation of this Dyna is expensive and in addition to the ZWGRU-Aktua lization errors recorded other effects that lead to a dyna lead to a knocking tendency. A clean, physika lisch based and thus reproducible separation of these Dynamic effects are therefore not possible.

ADVANTAGES OF THE INVENTION

The inventive method with the features of the claim 1 and the corresponding device according to claim 8 have the advantage over the known approaches Share that they are a physically based, dynamic precise tracking of the ignition parameters (ignition angle, closing angle).

In improvement of the prior art, it is proposed a dynamic correction of the ignition parameters on the Level of the load signal or filling signal in the form of a  Make a prediction calculation. This means that in Dy namikfall instead of one the instant filling signal indicating a predicted signal as an input variable ß for the or the load-dependent maps of the ignition parameter is used.

In particular, it is suggested that at least the solid Ignition angle map, which the pilot ignition angle under optimal conditions, and furthermore optio nal also other maps, such as B. the stationary adaptation map of the knock control, the load-dependent Maps of the closing angle calculation etc. on the predicate graced signal.

It becomes a dynamically more precise, already existing one Fill signal used so that the previously required Corrections at the ignition angle level (e.g. adaptation of an ignition angle dynamic reserve in the knock control) only in less (and thus with reduced functionality and Computing time) are necessary.

This becomes a major cause of dynamic knock, which due to the dynamic adaptation of the knocking frequency used today can not be taken into account optimally.

Advantageous further developments can be found in the subclaims improvements and improvements.  

According to a preferred development, the future Load signal predicted from a current main load si gnal, a current auxiliary load signal that corresponds to the current Main load signal leads ahead, and a crank angle inter vall that depends on the in time units or crank win calculation units can be specified.

According to a further preferred development, the ak actual auxiliary load signal from the opening angle of the throttle flap or the angle of the accelerator pedal, the speed of the Internal combustion engine and one if necessary by a Bypass channel to the throttle valve and / or by additional Bypass valves determined the amount of air flowing.

According to a further preferred development, the ak actual main load signal from the measured intake manifold pressure and the speed from which measured with an air mass meter Air mass or by filtering the current auxiliary load si gnals determined.

According to a further preferred development, a Providing a characteristic field for the ignition point depending on the load signal and the speed.

According to a further preferred development, this is done Predictions of the future load signal taking into account the camshaft adjustment and / or the exhaust gas recirculation tion.  

According to a further preferred development, white Other maps, such as in particular the stationary adaptation map of the knock control or one or more load-dependent current maps of the closing angle calculation, ready sets and takes place the respective ignition time point from the relevant characteristic curve field under Basis setting of the predicted future load signal.

DRAWINGS

Embodiments of the invention are in the drawings shown and in the description below he purifies.

Show it:

Fig. 1 is a flowchart for the basic flow of an embodiment of the method according to the invention;

FIG. 2 shows the technical background of an internal combustion engine in which the present invention is applicable;

Fig. 3 shows the time relationship between the load and the Zündzeitpunktberechnung ignition and the opening degree of the intake valve of a cylinder as a function of crank angle;

Figure 4 shows the course of the main load signal and the auxiliary load signal depending on the crank angle.

Fig. 5 shows the basic relationship between knock limit and optimal ignition and

Fig. 6 is a schematic representation of a characteristic field (KF) for the ignition timing or Zündwin angle.

DESCRIPTION OF THE EMBODIMENTS

In the figures, the same reference symbols designate the same or functionally identical components.

Fig. 3 shows the technical background of an internal combustion engine 100 in which the present invention is applicable.

First, the illustrated components for controlling internal combustion engine 100 are explained in more detail. An air / fuel mixture is supplied to the internal combustion engine 100 via an intake tract 102 , and the exhaust gases are discharged into an exhaust gas duct 104 . In the intake tract 102 , seen in the flow direction of the intake air, an air flow meter or air mass meter 106 , for example a hot-wire air mass meter, a temperature sensor 108 for detecting the intake air temperature, a throttle valve 110 with a sensor 111 for detecting the opening angle of the throttle valve 110 , a pressure sensor 112 and one or more a spray nozzle 113 attached. As a rule, the airflow meter or airflow meter 106 and the pressure sensor 112 are alternatively present.

A bypass channel 114 , in which an idle actuator 115 is arranged, leads around the throttle valve 110 . The bypass channel 114 and the idle actuator 115 can be omitted if the idle speed is controlled by means of the throttle valve 110 . If necessary, there may also be bypass valves that ensure sufficient idle speed when switching on an air conditioning system, for example. In the exhaust duct 104 , an oxygen sensor 116 is attached. On the internal combustion engine 100 , a crank angle sensor 118 and a sensor 119 for detecting the temperature of the internal combustion engine 100 are introduced . Furthermore, the internal combustion engine 100 has four spark plugs 120 for ignition of the air / fuel mixture in the cylinders, for example.

The output signals of the sensors described are transmitted to a central control unit 122 . Specifically, these are the following signals: a signal m of the air flow meter or air mass meter 106 , a signal T of the temperature sensor 108 for detecting the intake air temperature, a signal α of the sensor 111 for detecting the opening angle of the throttle valve 110 , a signal p of the pressure sensor 112 , a signal λ of the oxygen sensor 116 , a signal w of the crank angle sensor 118 and a signal TBKM of the sensor 119 for detecting the temperature of the internal combustion engine 100 . The control unit 122 evaluates the sensor signals and controls the injection nozzle or injectors 113 , the idle actuator 115 and the spark plugs 120 .

Fig. 3 shows the relationship between the load signal tL, the ignition angle calculation and ignition and the degree of opening of the intake valve of a cylinder as a function of the cure angle w or time t for an internal combustion engine with four cylinders.

The lower solid line shows the degree of opening of the Intake valve for cylinder No. 4 on, the lower stri smiled line the opening degree of the inlet valves of the rest Cylinder. It starts with a crank angle w of 0 ° Open intake valve of cylinder No. 4. At w = 90 ° the inlet valve of cylinder No. 4 opened to the maximum, and at w = 180 ° the inlet valve is closed again. There after passing through the intake valves of cylinders No. 2, No. 1 and No. 3 in this order the same opening and closing cycle and at a crank angle w of 720 ° be the inlet valve of cylinder No. 4 starts to close again to open.

The upper curve of Fig. 3 shows the course of the main load signal tL. The main load signal tL can be determined, for example, from the signal p generated by the pressure sensor 112 and the rotational speed n or from the averaged and filtered signal m of the air mass meter 106 .

The air filling and thus the load signal tL of the corresponding cylinder (in FIG. 4 cylinder No. 4) is required to calculate the ignition point. The main load signal tL at a certain crank angle in the vicinity of the crank angle at which the intake valve of the cylinder closes (in Fig. 4 cylinder No. 4, about 20 ° before the intake valve closes at 900 °) is representative of the air filling lung. This specific crank angle is referred to below as the filling angle. The exact value of the filling angle depends on the type of internal combustion engine 100 and can be determined empirically, for example.

Also indicated on the time axis are the ignition point t _z (crank angle w _z ), the closing point (t _s ) and a calculation time interval t _R for calculating the ignition point.

As described above and shown in Fig. 3, the calculation of the ignition timing at the closing time t _s must be completed, that is, long before the filling angle, which is reached at the time t ₀ + Δt.

For the calculation of the ignition timing, however, the air filling is used, which is represented by the main load signal tL present at the filling angle. However, the future course of the main load signal tL is generally not known, since it depends, for example, on the driver's request. If the main load signal tL or the average main load signal tLM used at the time of the calculation is used in the calculation, this leads to a non-optimal ignition timing if the main load signal tL changes until the filling angle is reached (see upper curve in FIG. 3 ), ie in non-stationary operation.

The method known from DE 44 01 828 A1 enables an approximate prediction of the load signal tL present at the filling angle, which is referred to below as the future load signal tLPr. It is particularly exploited that the main influencing factor on the course of the future load signal tLPr, the opening angle α of the throttle valve 111 , is known and that the signal α leads the signal tL a lot. Further details are shown in FIG. 4.

Fig. 4 shows a diagram in which the main load signal tL (dashed line) and the auxiliary load signal tL '(by a solid line) are plotted against the crank angle w. In stationary operation, the curves for tL and tL 'coincide (left to right). At the transition from low to high load, the curve for tL 'rises much faster than the curve for tL, so that future values for tL can be predicted from current values for tL' and tL, ie from the current auxiliary load signal tL 'and the current main load signal tL, the future load signal tLPr can be determined.

A simple intake manifold model can be used as a basis for determining the future load signal tLPr, which is described by a low-pass filter of the first order with a load-dependent filter constant. At the current crank angle w, the future load signal tLPr present at the future crank angle w + wPr is predicted according to the following equation:

tLPr = tL (w + wPr) =
tL (w) + (tL '(w) -tL (w)). (l-exp (-wPr / wF))

WPr is the prediction angle, i.e. the difference from the future crank angle for which the future Load signal tLPr is predicted - usually this is the filling angle - and the current crank angle w.

A flow chart of the basic sequence of an embodiment of Fig. 1 shows the inventive method.

In a first step 400 , the prediction angle wPr is determined. Step 400 is followed by step 402 , in which the auxiliary load signal tL 'is determined. The auxiliary load signal tL 'is determined as a function of the throttle valve angle α, the speed n and, if appropriate, the air volume flowing through a bypass channel 114 and / or additional bypass valves qLL from a characteristic diagram. Step 402 is followed by step 404 . In step 404 , the current main load signal tL is determined. The instantaneous main load signal tL can be determined, for example, by filtering the measured air mass m averaged over a crank angle segment with a low-pass filter of the first order. Alternatively, the current main load signal tL can also be determined from the intake manifold pressure p and the speed n or by filtering the auxiliary load signal tL '. Step 404 is followed by step 406 , in which the load-dependent filter constant wF is determined. This is followed by step 408 . In step 408 , the future load signal tLPr = tL (w + wPr) for the crank angle w + wPr, ie here the filling angle, is determined from the variables determined in steps 400 to 406 according to the equation mentioned above.

Ultimately, the ignition timing ZWGRU is determined in step 410 from the characteristic field KF on the basis of the predicted future load signal tLPr and the engine speed n. Then the flow of the flow diagram is ended.

Although the present invention has been described above in terms of preferred embodiment has been described, it is not limited to this, but in a variety of ways identifiable.

In particular, the characteristic field for the ignition timing also provided depending on the degree of filling or get saved. In this case there would be an intermediate  step to convert the detected load signal into the fill degree of efficiency necessary.

Nor is the invention exemplary of the above Prediction procedure limited. For example, the Ak Update error of the ignition angle calculation by the ver application of a predicted signal under additional Taking camshaft adjustment and exhaust back into account leadership can be compensated.  

REFERENCE SIGN LIST

wPr prediction angle
tL main load signal
tL 'auxiliary load signal
w crank angle
t time
wF filter constant
tLPr predicted future load signal
ZWGRU basic ignition angle
KF map
n speed

100

Internal combustion engine

102

Intake tract

104

Exhaust duct

106

Air mass meter

108

Temperature sensor

110

throttle

111

Sensor for detecting the opening win throttle valve

110

112

Pressure sensor

113

Injectors

114

Bypass channel

115

Idle actuator

116

Oxygen sensor

118

Crank angle sensor

119

Temperature sensor

120

Spark plugs
t _s

Closing time
t _z

Ignition timing
tLM medium load signal
[t _-1

, t ₀

] Time interval for averaging
t _R

Calculation interval
w _z

Firing angle

Claims (8)

1. Method for determining the ignition timing of an internal combustion engine with the following steps:
Providing a characteristic field (KF) for the ignition point (t _z ) as a function of at least one of the relati ve air filling of a cylinder of the internal combustion engine to load signal (tL);
Detecting the load signal (tL) at a point in time (t ₀ ) before the ignition point (t _z );
Predictions of a future load signal (tLPr) at a later time, which is to be determined before the ignition point (t _z ) to be determined, lie in the point in time (t ₀ + Δ); and
Determining the ignition point (t _z ) from the characteristic field (KF) on the basis of the predicted future load signal (tLPr). 2. The method according to claim 1, characterized in that the future load signal (tLPr) is predicted from egg  a current main load signal (tL), a current auxiliary load signal (tL ') that is ahead of the current main load signal hurries, and a crank angle interval (wPr) that depends from that in time units or crank angle units pressed calculation time (wB) can be specified. 3. The method according to claim 2, characterized in that the current auxiliary load signal (tL ') from the opening angle (α) of the throttle valve ( 110 ), the speed (s) of the internal combustion engine ( 100 ) and optionally by a bypass channel ( 114 ) to the throttle valve ( 110 ) and / or the amount of air flowing through additional bypass valves (qLL) it is averaged. 4. The method according to claim 2 or 3, characterized in that the current main load signal (tL) from the measured intake manifold pressure (p) and the speed (n), from the egg mass meter ( 106 ) detected air mass (m) or is determined by filtering the current auxiliary load signal (tL '). 5. The method according to any one of the preceding claims, characterized in that a characteristic field (KF) is provided for the ignition timing (t _z ) as a function of the load signal (tL) and the speed (n). 6. The method according to any one of the preceding claims, since characterized by that predicting the future  Load signal (tLPr) taking into account the camshafts adjustment and / or exhaust gas recirculation. 7. The method according to any one of the preceding claims characterized in that further maps, such as the stationary adaptation map of the knock control or one or more load-dependent maps of the closing winch Calculation, provided and a setting of the respective ignition timing from the relevant characteristic curves field based on the predicted future Load signal occurs. 8. Device for determining the ignition timing of a racing engine with:
a memory device for providing a characteristic field (KF) for the ignition point (t _z ) as a function of at least one load signal (tL) indicating the relative air filling of a cylinder of the internal combustion engine;
a detection device for detecting the load signal (tL) at a time (t ₀ ) before the ignition point (t _z ) to be determined;
a prediction device for predicting a future load signal (tLPr) at a later point in time (t ₀ + Δ) before the ignition point (t _z ) to be determined; and a determining device for determining the ignition point (t _z ) from the characteristic field (KF) on the basis of the predicted future load signal (tLPr).