EP1646777B1 - Method and device for controlling an internal combustion engine - Google Patents

Abstract

Disclosed are a device and a method for controlling an internal combustion engine. A first variable characterizing the actually injected fuel quantity and a second variable characterizing the desired fuel quantity that is to be injected are determined based on operating characteristics. The first variable is compared to the second variable. A first correction value for correcting a fuel quantity and a second correction value for correcting the air quantity can be predefined based on said comparison, the first correction value being limited to a maximum value.

Description

State of the art

The invention relates to a method and a device for controlling an internal combustion engine according to the preambles of the independent claims.

From the DE 198 31 748 a method and a device for controlling an internal combustion engine is known. A correction unit calculates a correction value for the correction of a pump characteristic field on the basis of the fuel quantity to be injected and various input signals such as, for example, the air quantity and a lambda signal. In the correction device, the fuel quantity to be injected and the actually injected fuel quantity are compared and, depending on this comparison, the pump map is corrected such that the defined relationship between the fuel quantity to be injected and the control signal for the fuel pump is restored. It is further provided that, starting from a humidity signal, a correction signal for correcting the air quantity is specified. The desired amount of fuel QKW is limited to a maximum allowable amount of fuel through a minimum selection. This limited amount of fuel then serves as fuel to be injected.

A method and a device for controlling an internal combustion engine is known from the unpublished DE 102 21376 known. There, a method and a device for controlling an internal combustion engine is described in which, based on operating parameters, a lambda value of the exhaust gas is determined. This is compared with the actual lambda value and, based on the comparison, a correction value for the correction of a fuel quantity or an air quantity signal is calculated.

In essence, a first variable characterizing the actually injected fuel quantity is determined from the sensor signal of a lambda sensor and an air mass sensor and compared with a second variable characterizing the desired fuel quantity to be injected. Based on this comparison, a first correction value for the correction of a fuel quantity and / or a second correction value for the correction of an air quantity is specified.

In an ideal error-free system, the actual amount of fuel injected would have to correspond to the desired amount of fuel. Due to tolerances and / or aging effects, the case occurs that the desired fuel quantity deviates from the actually injected fuel quantity. If now the amount of air metered to the internal combustion engine is controlled and / or regulated as a function of the desired amount of fuel to be injected, a faulty amount of air is set. A control depending on the actually injected amount of fuel is not readily possible because it is difficult to detect. By measuring the lambda value of the exhaust gas and the amount of air supplied to the internal combustion engine, the actual injected fuel quantity can be calculated and compared with the desired amount of fuel to be injected. Based on the deviation of these two signals results in a correction value. With this correction value can now be intervened on the air system. This is done, for example, such that the fuel amount value supplied to the air system is corrected with the corresponding correction value. Furthermore, it can be provided that directly the amount of air is corrected accordingly. As an alternative to calculating the fuel quantity, the lambda signals or other quantities characterizing the fuel quantity can also be used directly.

Alternatively, it can also be provided that intervention is made directly in the fuel metering system in such a way that a quantity of fuel quantity is corrected by means of the correction value until the amount of fuel to be injected and the fuel quantity actually injected coincide. Such a direct correction of the amount of fuel is problematic because such a correction can lead to an increase in quantity. For safety reasons, therefore, it is not desirable that the direct intervention of the quantity corrects arbitrarily large deviations or acts in the entire engine operating range.

These restrictions do not apply to indirect intervention, for example via the air control system by means of exhaust gas recirculation. Since emissions are equivalent or better than indirect intervention, indirect intervention on the amount of air is usually preferred.

According to the invention, it has been recognized that errors in the injection quantity can possibly have a negative effect on the driving behavior.

According to the invention it is therefore provided that the correction value acts on the amount of fuel and / or on the amount of air. In this case, the correction value, which acts on the amount of fuel, limited to a maximum value. By means of this procedure, both effects on the exhaust emissions and on the driving behavior can be compensated. In a preferred embodiment it is provided that the entire error is compensated by means of a direct intervention. If this is not possible, the remaining error is compensated by means of an indirect intervention. The direct intervention affects the amount of fuel and the indirect intervention affects the amount of air.

According to the invention, the quantity error corresponding to the deviation between the actual and the desired fuel quantity is proportionately compensated for via a direct intervention in the metering and an adaptation to the air mass to the remaining quantity error.

It is particularly advantageous if the type of intervention is dependent on the engine operating state. This is realized, for example, in that the limit and thus the proportion of the direct intervention is specified as a function of operating conditions and is thus continuously adjusted. In this case, preferably the rotational speed and / or a variable characterizing the load of the internal combustion engine are used as operating parameters.

It is preferably provided that the first and / or the second correction value be adapted. That is to say, in states in which the correction values can be determined, the correction values are stored in one or more characteristic fields depending on the operating state of the internal combustion engine or quantities are determined and stored which can be used to calculate the correction values according to a mathematical method. In states in which the correction values can not be determined, the stored correction values or the stored variables are used.

In a particularly advantageous embodiment, it is provided that the cylinders of the internal combustion engine are divided into at least two groups, and that different second correction values are specified for the different groups. This means that the mean quantity error of the two groups is corrected by a fuel quantity intervention. The remaining and / or the individual errors of the individual groups are compensated by an indirect intervention.

It is preferably provided that the correction takes place by means of a fuel quantity intervention up to a certain error. In the case of larger and / or asymmetrical errors, an additional correction by means of an air quantity intervention takes place.

drawing

The invention will be explained below with reference to the embodiments shown in the drawing.

Figur 1

ein Blockdiagramm der erfindungsgemäßen Vorrichtung,

Figur 2

und

Figur3

jeweils eine Ausgestaltung für eine Brennkraftmaschine, bei der die Zylinder der Brennkraftmaschine in wenigstens zwei Gruppen aufgeteilt sind.

Show it

FIG. 1

a block diagram of the device according to the invention,

FIG. 2

and

Figur3

in each case an embodiment for an internal combustion engine, in which the cylinders of the internal combustion engine are divided into at least two groups.

The procedure according to the invention is described below using the example of the fuel quantity to be injected. Instead of the amount of fuel, other quantities that characterize the amount of fuel can be used. In particular, torque quantities, fuel volumes and / or the activation duration of corresponding actuators can be used.

In FIG. 1 is a fuel quantity control designated 100. This is dependent on various input variables, such as the speed of the internal combustion engine and a signal FP, which characterizes the driver's desire, a desired amount of fuel to be injected MES. This is also referred to below as the second size. This signal regarding the desired amount of fuel to be injected passes through a node 105 to a fuel quantity actuator 110. The fuel quantity actuator 110 determines the time and the end and thus the duration of the fuel metering. Preferably, this is designed as a solenoid valve or as a piezoelectric actuator, which is preferably arranged in an injector, an injection nozzle, or another actuator.

An air quantity control 200 supplies an air quantity signal MLS on the basis of various input variables, such as the engine speed N and a quantity MES characterizing the quantity of fuel to be injected. As input variable for the amount of fuel to be injected, the output signal of the quantity control 100 is preferably used. With the output signal MLS of the air quantity control 200, an air quantity actuator 210 is acted upon via a node 205. Depending on the signal MLS with respect to the desired amount of fresh air, the air quantity actuator 210 sets the appropriate amount of air. This is preferably an actuator for influencing the amount of recirculated exhaust gas in the form of an exhaust gas recirculation controller, a throttle valve, which influences the amount of air supplied to the internal combustion engine, and / or a supercharger.

A fuel quantity calculation 120 determines, based on various input variables, a quantity MEI which characterizes the actually injected fuel quantity, which is also referred to below as the first quantity. As an input variable, the fuel quantity calculation processes, in particular, a signal L, which characterizes the oxygen concentration in the exhaust gas, and a signal MLI, which characterizes the air quantity supplied to the internal combustion engine. The two signals are preferably provided by sensors, in particular a lambda probe and an air mass meter. Alternatively, these signals can also be determined on the basis of other variables.

In addition to the in FIG. 1 Input variables can be taken into account by the fuel quantity control, the air flow control and the fuel quantity calculation yet other input variables.

The first and the second quantities MES and MEI arrive at a node 125 with different signs. The output point DME of the node indicates the deviation between the actually injected fuel quantity and the desired quantity of fuel to be injected. This signal DME with respect to the injection quantity error passes via an integrator 130 and a limiter 132 to a first characteristic map 134. The output QME of the first characteristic field is applied to the second input of the connection point 105. The limiter 132 in turn supplies the integrator 130 with a signal. Both the limiter 132 and the map 134 are different Operating characteristics, such as the speed N of the internal combustion engine and other variables supplied.

Furthermore, the signal DME with respect to the injection amount error passes through a filter 140 and a sign inverter 142 to a second map 144, whose output QML of the second input of the node 205 is applied. The second map 144 also different signals with respect to various operating characteristics such as the rotational speed N are supplied.

The integrator 130 and the limiter 132 act as integral controllers with output limiting and anti-windup function. That is, the injection quantity error is integrated by the integrator 130. Upon reaching the limit value of the limiter 132, the integrator is stopped, this is indicated by the connection between the limiter and the integrator 130. Once the limit value of the limiter 132 is reached, the output of the limiter remains at the value achieved.

The limiting value of the limiter 132, to which the output signal of the integrator 130 is limited, according to the invention in one embodiment depending on the operating condition of the internal combustion engine can be predetermined. The limiting value is preferably predefined as a function of the rotational speed N of the internal combustion engine and / or further operating parameters.

The output of limiter 132 is the amount of error that is to be compensated for by direct intervention on the amount of fuel. This is adapted in the subsequent first map 134. This means, if a specific operating point of the internal combustion engine is reached, which is preferably defined by the rotational speed and the load, then based on the comparison between the first and the second variable, the injection quantity error is determined and integrated and limited. The value thus determined is then stored in the map 134 depending on the operating point.

According to the invention, it is now provided that a correction of the fuel quantity should take place only in certain operating ranges. This is ensured by the fact that in the other operating areas, in which no fuel quantity correction is to take place Limit value is set to zero. In the other operating points, the fuel metering and thus the driving behavior is adapted. In the other operating points or in operating points in which the limiter is active, ie the error can not be completely corrected by the fuel quantity correction, an additional correction of the amount of air. That is, it is either corrected only the amount of fuel or just the amount of air or both amounts are corrected.

This means that the limit can be adjusted continuously for different operating points. The remaining quantity error is automatically compensated by the amount of air.

If the integrator reaches the limit, the injection amount error will not be completely corrected via the fuel metering. Accordingly, the input signal of the integrator remains nonzero, i. the injection quantity error is not equal to zero. This remaining injection quantity error is compensated by the amount of air. The signs of the two interventions differ, this is ensured by the inverted 142. Via the filter 140, which is preferably realized as a low-pass filter, the dynamics of the air branch can be applied independently of the fuel quantity measurement. Preferably, the air flow branch has a dynamically slower behavior so that the learning of the fuel quantity correction is not unnecessarily affected.

At operating points in which the first quantity MEI is known, the correction quantities QME for the quantity of fuel to be injected and QML for the amount of air are calculated and stored in the maps 134 and 144 depending on the respective operating point, i. learned. If the first variable MEI does not exist, this is the case, for example, if the lambda signal does not provide reliable values, the values stored in the maps 134 and 144 are used to correct the fuel quantity and / or the air quantity.

Instead of the maps 134 and 144, other learning functions or adaptive methods can be used.

In the FIG. 2 a further embodiment of the procedure according to the invention is shown. This procedure is provided in particular for special so-called V-engines, which essentially consist of two in-line engines which have a common crankshaft. However, this embodiment is not limited to such engines, it is generally applicable to internal combustion engines, in which the cylinders of the internal combustion engine are assigned to different banks / groups, wherein each of the banks / groups is assigned in each case an actuating element for influencing the amount of air.
Furthermore, the approach is also applicable to a larger number of banks. In particular, the procedure can also be used if each cylinder is assigned an actuating element for influencing the amount of air.

Already in FIG. 1 Elements described are designated by corresponding reference numerals. In essence, the design of the differs FIG. 2 of the FIG. 1 in that two quantity calculations 120 are provided for the amount of fuel actually injected. The quantity calculation for the first bank is the same as in FIG. 1 designated. The quantity calculation for the second bank is designated 320. The first variable associated with the first bank is hereinafter referred to as MEIL and the first variable associated with the second bank is designated MEIR. The node 125 of the first bank corresponds to the node 325 of the second bank. The quantity error of the first bank is denoted by DMEL and the quantity error of the second bank by DMER. The first bank elements 140, 142, 144 and 205 are labeled 340, 342, 344 and 305 at the second bank. The operation of these elements corresponds to the operation of the corresponding elements of the FIG. 1 ,

The integrator 130 is supplied with the output of a divider 350, which processes the output of the link 160. The node 160 is supplied with the injection quantity error of the first bank DMEL and the injection amount error of the second bank DMER. This means that the integrator is supplied with the mean value of the two injection quantity errors of the two different banks. It is understood that the input signals of the quantity calculation 120 or 320 are provided by different sensors that are assigned to the individual banks.

According to the invention, it is now provided that the procedure of FIG. 1 is essentially transferred to one of the banks, ie the individual elements are designed twice. The correction of the amount of fuel is uniform for both banks. This is necessary because a different correction would cause interference with other controls. If the limit is reached during the fuel quantity correction, the remaining bank-specific residual errors are compensated via the air volume interventions. The same applies if different injection quantity errors occur for the different banks. In this case, the average error is compensated by the fuel quantity intervention, and the bank-individual residual errors are additionally compensated by the air volume interventions.

In the FIG. 3 another embodiment is shown. It essentially corresponds to the functionality of the embodiment 2, but requires less overhead on computer runtime and storage space requirements. Already in FIG. 2 and 1 Elements described are designated by corresponding reference numerals. The injection quantity error DMEL of the first bank arrives at a connection point 410 and at a connection point 420. Accordingly, the injection quantity error of the second bank DMER likewise reaches the two connection points 410 and 420. In the connection point 410, the sum of the two signals and, in the connection point 420, the difference between the two formed two signals. In the subsequent division means 415 and 425, the output signals of the connection points 410 and 420 are divided by two. The filter 140 is thus supplied with the mean value of the two injection quantity errors of the two banks. The filter 340 is supplied with the deviation from the mean value. With the output signal of the map 144 on the one hand, a filter 430 and on the other the two nodes 440 and 450 applied. The filter is preferably designed as a factor member. Accordingly, the output of the map 344, the two nodes 440 and 450 are applied. The output signal of the filter 430 reaches the limiter 132. The signal QMLL is present at the output of the connection point 440 and the signal QMLR is present at the output of the connection point 450.

According to the invention, in this embodiment, the mean value and the half difference, ie the deviation from the mean value of the individual errors in the maps 144 or 344 learned. From these quantities, the three correction terms QME, QMLL and QMLR are determined by suitable adaptive linkage with suitable sign choice. That is, the elements 430 and 132 are predeterminable depending on the operating point. The two interventions on the air quantity are symmetrical with respect to the mean with the opposite sign. The maps 144 and / or 344 may alternatively be configured as any learning functions.

For learning the mean value, no integrator but a low-pass filter 140 is used. For this reason, the quantity error is never completely compensated by the intervention on the amount of fuel. It always acts at the same time an intervention on the amount of air. The transmission behavior of the filter 430, as well as the values of the limits of the limiter 132, can be predetermined depending on the operating state.

In the embodiments of the Figures 2 and 3 the correction takes place via a uniform intervention on the fuel quantity for all cylinders. The correction by means of the intervention on the air quantity takes place individually for different groups of cylinders. It can be provided that the correction takes place for individual cylinders or common to several cylinders. The number of correction values preferably corresponds to the number of air mass meters and / or the number of control elements.

Claims (7)

1. Method for controlling an internal combustion engine, in which method a first variable, which characterizes the actually injected fuel quantity, and a second variable, which characterizes the desired fuel quantity to be injected, are determined on the basis of characteristic operating variables, and the first variable is compared with the second variable, with a first corrective value for correcting a fuel quantity being predefined on the basis of the comparison of the first variable with the second variable, with a second corrective value for correcting an air quantity being predefined on the basis of the comparison of the first variable with the second variable, and with the first corrective value being limited to a maximum value.
2. Method according to one of the preceding claims, characterized in that the first and/or the second corrective value are adapted.
3. Method according to one of the preceding claims, characterized in that the maximum value can be predefined as a function of characteristic operating variables.
4. Method according to one of the preceding claims, characterized in that the first and/or the second corrective value is stored as a function of characteristic operating variables.
5. Method according to one of the preceding claims, characterized in that the second corrective value is delayed in terms of time with respect to the first corrective value.
6. Method according to one of the preceding claims, characterized in that the cylinders of the internal combustion engine are divided into at least two groups, and in that different second corrective values are predefined for the different groups.
7. Device for controlling an internal combustion engine, having means which determine a first variable, which characterizes the actually injected fuel quantity, and a second variable, which characterizes the desired fuel quantity to be injected, on the basis of characteristic operating variables, predefine a first corrective value, for correcting a fuel quantity, on the basis of the comparison of the first variable with the second variable, predefine a second corrective value, for correcting an air quantity, on the basis of the comparison of the first variable with the second variable, and limit the first corrective value to a maximum value.

Images