EP0579794A1

EP0579794A1 - Process and device for assessing the efficiency of a lambda control system.

Info

Publication number: EP0579794A1
Application number: EP93901632A
Authority: EP
Inventors: Ernst Wild
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1992-02-07
Filing date: 1993-01-14
Publication date: 1994-01-26
Anticipated expiration: 2013-01-14
Also published as: WO1993016277A1; KR100237272B1; DE59301082D1; EP0579794B1; DE4203502A1; JPH06506752A; JP3451087B2; US5404861A

Abstract

Procédé permettant d'évaluer la capacité à fonctionner d'une régulation lambda qui fournit des valeurs de positionnement FR devant varier autour d'une valeur de positionnement théorique FR-SOLL. Cette régulation lambda est assistée par un dispositif d'adaptation qui fournit des données d'adaptation. Ce procédé se caractérise par le fait: que la valeur actuelle EW d'une grandeur de décision est calculée en continu; cette grandeur de décision indique l'écart moyen des valeurs de positionnement par rapport à la valeur de positionnement théorique; que la valeur actuelle est comparée à une valeur de seuil des grandeurs de décision SW-EW; et qu'un signal d'erreur est émis lorsque la valeur actuelle excède la valeur de seuil des grandeurs de décision. Ce procédé présente l'avantage de permettre également d'identifier des erreurs qui ne se produisent que dans des parties de l'ensemble de la plage de fonctionnement d'un moteur à combustion interne, pour laquelle l'alimentation en carburant est régulée par la régulation lambda. Lorsqu'un cas de ce type se produit, la valeur de positionnement fournie par la régulation lambda s'écarte de la valeur de positionnement théorique, ce qui entraîne la modification d'une valeur d'adaptation. Si la plage générant des erreurs est à nouveau quittée, la valeur d'adaptation modifiée ne convient alors plus à la plage de fonctionnement correct. C'est la raison pour laquelle la valeur de positionnement fournie par la régulation lambda s'écarte dès lors de la valeur de positionnement théorique, dans l'autre sens. Comme on établit la moyenne de ces écarts lors du calcul de la valeur de décision, ils interviennent davantage dans la valeur de décision que dans la (ou du moins d'une) valeur d'adaptation qui est à nouveau abaissée dès que le signe de l'écart de la valeur de positionnement s'est inversé. La valeur de décision permet de déterminer des erreurs qu'une valeur d'adaptation ne permet pas de déceler.A method of evaluating the ability to operate of a lambda regulation which provides positioning values FR which must vary around a theoretical positioning value FR-SOLL. This lambda regulation is assisted by an adaptation device which supplies adaptation data. This method is characterized by the fact that: the current value EW of a decision quantity is calculated continuously; this decision quantity indicates the average deviation of the positioning values from the theoretical positioning value; that the current value is compared to a threshold value of the decision quantities SW-EW; and that an error signal is output when the current value exceeds the threshold value of the decision quantities. This method has the advantage of also making it possible to identify errors which only occur in parts of the entire operating range of an internal combustion engine, for which the fuel supply is regulated by the engine. lambda regulation. When such a case occurs, the positioning value supplied by the lambda control deviates from the theoretical positioning value, which leads to the modification of an adaptation value. If the error-causing range is left again, then the changed adaptation value no longer matches the correct operating range. This is the reason why the positioning value supplied by the lambda regulation therefore deviates from the theoretical positioning value, in the other direction. As these deviations are averaged when calculating the decision value, they are more involved in the decision value than in the (or at least one) adaptation value which is lowered again as soon as the sign of the deviation of the positioning value has reversed. The decision value makes it possible to determine errors that an adaptation value cannot detect.

Description

Method and device for assessing the operability of a lambda control

The following relates to a method and a device for assessing the operability of a lambda control for an internal combustion engine, which outputs control values FR which are intended to fluctuate by a target value FR TARGET.

State of the art

In order to generate as little harmful gas as possible when operating internal combustion engines, they are operated with a lambda control with pilot control. As a result, fuel tightnesses are determined, which are to be supplied to the engine in such a way that they match the intake air in such a way that a predetermined lambda value is maintained as precisely as possible. If values of operating variables change, the pilot control immediately determines a value for the fuel metering which is adapted to the changed operating values and which is then finely adjusted with the aid of the lambda control.

The pilot control values are determined for a respective internal combustion engine for precisely defined operating values and operating parameters. Now, however, the current operating conditions give way to the practical operation of an internal combustion engine. drive parameters often from those used in determining the pilot values, e.g. E. a different fuel is used. Then the predetermined pre-control values do not exactly match the current operating case. To remedy this deficiency, there are so-called learning or adaptive lambda control systems. These output at least one adaptation value, with the aid of which the pilot control values are corrected. The adaptation value is determined with the aid of the deviation that the control value output by the lambda controller has from a target control value.

During the operation of an internal combustion engine. the. Errors increasing harmful gas emissions occur. The California environmental agency CARB demands that an error should be displayed if the permissible limit value for a harmful gas is exceeded by 50% in the so-called AGV cycle. In this context, it proposed to monitor at least one adaptation value and to output an error signal if it exceeds a predetermined threshold value.

It has been found that methods and devices according to this proposal are not able to display all errors which lead to the limit value for a harmful gas being exceeded by 50% in the FTP cycle.

The object was accordingly to specify a method and a device for assessing the operability of a lambda control, which are able to indicate difficulties in the control which lead to an undesirable increase in the emission of harmful gas.

Presentation of the inventions

The inventive method for assessing the functional Ability of a lambda control that outputs control values FR that are to fluctuate by a target value FR_SOLL, which lambda control is supported by an adaptation that outputs adaptation values, is characterized in that

- The current value EW of a decision variable is continuously calculated, which indicates the average amount of deviation of the manipulated values from the setpoint;

- The current value is compared with a decision variable threshold value SW_EW; and

- an error signal is issued if the current value. exceeds the decision size threshold.

Preferably, not only the decision values are used to assess whether the error signal is to be output, but also the values of the at least one adaptation variable are used. In this case, the error signal is output either when the current decision value exceeds the associated threshold value or when an adaptation value exceeds its associated threshold value. In this case, where the decision value is used in addition to the adaptation values for assessing the functionality, it is further advantageous to determine the decision value with a larger time constant than the at least one adaptation value. Then errors are generally displayed via the adaptation values, while the decision value is only displayed in special cases.

The knowledge on which the above-mentioned method is based is now illustrated using an example. It is assumed that at high loads the fuel pump on the lambda-controlled internal combustion engine can no longer deliver the requested amount of fuel. Then a lean air / fuel mixture appears. The result of this is that the set value of the sol 1 set value output by the lambda control _ Λ -

deviates. As a result, the values of the decision size and the adaptation sizes increase. After a maximum of a few 10 seconds, the high load range will be left. Because at least one adaptation value has been increased, a rich mixture is now set, as a result of which the manipulated value deviates from the setpoint manipulated value on the other side than before. The at least one increased adaptation value is therefore reduced again. On the other hand, the decision value is further increased since, in contrast to the adaptation values, what matters is the average deviation of the manipulated values from the target value in terms of the amount.

With the aid of the adaptation values, only errors which have an effect in the entire operating range of an internal combustion engine can be displayed, be it a little more or a little less in individual sub-areas. On the other hand, with the help of the decision value, not only are these errors known, but also those are recognized which only have an effect in a partial area.

The device according to the invention for assessing the functionality of a lambda control, which outputs control values FR, which should fluctuate by a target value FR_S0LL, which lambda control is supported by an adaptation that outputs adaptation values, is characterized by:

- A calculation device for continuously calculating the current value EW of a decision variable, which indicates the averaged amount of deviation of the manipulated values from the setpoint value; and

a comparison device which compares the current value with a decision variable threshold value SW_EW and outputs an error signal if the current value exceeds the decision variable threshold value. Drawing

1: Block function diagram for explaining a method according to the invention and a device according to the invention for assessing the functionality of a lambda control; and

2: Flow chart for describing a method for assessing the functionality of a lambda control.

Description of exemplary embodiments.

The block diagram of FIG. 1 shows an internal combustion engine 10 with a lambda control block 11 with pilot-controlled adaptive lambda control and an error message block 12.

In the lambda control block 11 there are a pilot control map 13, a lambda control 14, an adaptation 15, an adaptation adder 16, an adaptation multiplier 17 and a control multiplier 18. The pilot control map 13 is addressed via speed values n and load values L and outputs pilot values tv for injection times. An adaptive adaptation value AWA is added to a respective pilot control value in the adaptation adder 16, then multiplication is carried out in the adaptation multiplier 17 by a multiplicative adaptation value AWM, and finally multiplication in the regulation multiplier 18 by a regulation factor FR. The latter is formed by the lambda control 14 on the basis of a control deviation between an actual lambda value λ_IST and a sol 1 lambda value λ_SOLL. The control factor FR is the manipulated variable of the lambda control. The setpoint "1" is subtracted from this manipulated value in a subtraction device 19, and the adaptation values AWA and AWM are calculated by the adaptation 15 with the aid of the manipulated value deviation ΔFR thus formed. It is pointed out that in practice there are numerous variants of lambda control blocks, but all of them have essentially the same function as the one explained above. In this way, the pre-control values tv can be determined in a variety of ways, e.g. E. also without map. The adaptation adder 16 and the adaptation multiplier 17 can also be located behind the control multiplier 18 instead of in front of it. Instead of two adaptation values, the adaptation 15 can also output only a single or even three safe values or even more. In this way, at low speed and high load, leakage errors can be adapted, which are preferably additive before the link to the derr. Stεi lwert are taken into account. Multiplicative errors, such as those caused by changes in air pressure or changes in fuel properties, can be taken into account in a multiplicative manner behind the link with the set value. Finally, opening and shooting times can. vcr. Injection valves are adapted at high speed and high load and additively after the link with the. Actual value are taken into account.

The error reporting block 12 contains a calculation sheet 20 and a comparison block 21. The calculation device 20 receives the Ste11 value deviation ΔFR and uses this to calculate an expected value EW, preferably as a variance, i. H. as the mean of the squares of the deviation from the ste1, i.e. as:

EW = (ΔF ^** PR ^* ). 2

Instead of the variance, however, the simple absolute mean value can also be calculated as a decision value, i.e.:

EW ^* = i ΔFRl. Furthermore, an event number can be used as a decision value, for example the number that indicates how often within a predetermined period of time or within a predetermined number of examined manipulated value deviations the value j ΔFR I exceeds a threshold value, that is:

EW = frequency of j ΔFR "threshold.

For the determination of the decision value EW, it is not the specific type of calculation that is essential, but it is important that a deviation averaged in terms of amount be determined. It depends on the use of the amount kcmτ ^* .t so that set value deviations, such as those caused by the occurrence of an error, as well as those caused by the disappearance of this error, are taken into account. The averaging is of significance, so that not every rapidly passing larger value deviation for the output of an error signal FΞ occurs by the comparison block 21, namely the current one. The decision value EW is compared with a decision value threshold value and outputs the error signal when the decision value exceeds the decision size threshold value. In the exemplary embodiment, the averaging is carried out with the aid of a digital low-pass filter, as explained below with reference to step ε2 of the flowchart in FIG. 2. A low-pass constant is used which corresponds to a time constant of a few 10 seconds in the case of a corresponding integrating element.

A method can be carried out with the functional blocks according to FIG. 1, as will now be described with reference to FIG. 2.

After the start of the method in FIG. 2, the expected value EW is set to "1" in an initialization step si. In addition, threshold values SW_EW, SW_AWA and SW_AWM are set up predefined values set. In all three cases, the exemplary embodiment has the value 1.2.

The method then enters a loop in which the current one is carried out in a step S1. Values ΔFR of the manipulated variable deviation, AWA of the adaptive adaptation variable and AWM of the multiplicative adaptation variable are recorded. In the subsequent step 2 2 already mentioned above, the maintenance value EW is determined by digital low-pass filters r.σ from the previously valid value EW and the current manipulated variable deviation ΔFR with that specified in the block for step s2. Formula calculated. Therein, c is the low-pass constant which, for the exemplary embodiment, has the value 0.99 on ice.

Now decision steps s: __ to sr, i dene_. ;.-_ Is queried one after the other, C D the values A, A M or EW are each larger than the assigned threshold value SW _.- W.-. SW_AWM or SW_EW. Will not be one of those questioners. if yes, a final step se is examined, c an end condition is fulfilled. If so, the process will. aborted, otherwise the loop from step sl

in steps ε3 to s5, values are exceeded, in a step ε5 the lεr is entered in an error memory, and an error signal is output, which may e.g. B. brings a warning lamp to light up. After step so the end of the procedure is reached.

The method just described can be modified in many ways, as long as it is only ensured that it is examined whether the manipulated variable deviation ΔFR averaged exceeds an associated threshold value. In this way, the comparisons made with the aid of the adaptation values can be omitted entirely. Furthermore, a variant It is possible that the end of the procedure is not reached after the error message step, but that the loop mentioned is repeated from step S1 despite the error that has been found, thereby opening up a possibility of error recovery, e.g. E. in that the error entry is deleted again if an error does not occur again after a predetermined high number of runs. If the error message was triggered by the decision value exceeding its assigned threshold value, this error can also be used to store the values of selected operating variables as they existed when the error occurred. If dar ..: The same operating state is started several times without an error message again, the error entry can be deleted again.

It is essential for the method and the device according to the invention that when a region with incorrect control is approached, the manipulated value output by the lambda control deviates from the target value, which means that at least one adaptation value and the decision value change. When the defective portion of the wie¬ leave the modified adaptation value does not fit _ι more to error-free Eereich now therefore the target value outputted from the regeiung Lambda¬ in the other direction deviates from the desired regulated value. Since these deviations are averaged in terms of amount when calculating the decision value, they have a stronger effect on the decision value than on the at least one adaptation value, which is immediately reduced again as soon as the sign of the deviation in the set value has reversed. Therefore, with the help of the decision value, errors can be determined that cannot be detected with the help of an adaptation value.

Claims

1.Procedure to assess the operability of a lambda control that outputs control values FE that are one. Should the first value FR_SOLL fluctuate, which lambda control is supported by an adaptation that outputs adaptation values, characterized in that

- the current value EW of a decision variable is continuously calculated, which indicates the average deviation of the manipulated values from the target value;

- The current value is compared with a decision variable. SW11E value; and

- an error signal (FS) is output when the current value exceeds the decision size threshold (EW> SW_EW).

2. The method according to claim 1, characterized in that the decision values EW are calculated as follows:

EW = I (FR - FR TARGET)

3. The method according to claim 1, characterized in that the decision values are calculated as follows:

EW = (FR - FR_SOLL) ²

4. The method according to claim 1, characterized in that the decision values EW are calculated as follows:

EW = frequency of j (FR - FR_SOLL):.

5. The method according to any one of claims 1 to 4, characterized ge indicates that the averaging is carried out by digital low pass filter.

6. The method according to any one of claims I to 5, characterized ge indicates that the error signal (FS) is also output when at least one adaptation value exceeds an assigned adaptation threshold value (SW_AWA, SW_AWM).

7. The method ^'according to claim 6, characterized in that the expected value (Ξ) is determined with a greater time constant than the at least one adaptation value (AWA, AWM).

8. Device for assessing the functionality of a lambda control, the control values F? outputs that should fluctuate by a sol 1 set value FR_SOLL, which Lambdaege¬ iung is supported by an adaptation that outputs adaptation values, characterized by:

- A calculation device (20) for continuously calculating the current value EW of a decision variable, which indicates the averaged amount of deviation of the manipulated values from the desired manipulated value; and

- A comparison device (21) which compares the current value with a decision variable threshold value SW_EW and outputs an error signal when the current value exceeds the decision variable threshold value.