DE10339063A1 - Mixture regulation method for an internal combustion engine, involves calculating value of oxygen charging, and determining rich or poor breakthrough by oxygen sensor - Google Patents

Abstract

The mixture exchange is dependent on the calculated value of oxygen charging. A rich or poor breakthrough determined by oxygen sensor is initiated to the catalytic converter. The adapted value of the oxygen storage capability, changes dynamically depending on optimization parameters. An independent claim is also included for a device for mixture regulation of an internal combustion engine.

Description

The The invention relates to a method and a device for mixture control according to the generic terms of the independent Claims.

Lots Catalysts for Internal combustion engines have the property of storing oxygen. This memory effect allows a permanent implementation of pollutants and thus improves the Conversion properties, since in excess of oxygen (lean engine operation) this is stored and at reducing agent excess (rich engine operation) Some pollutants react with the stored oxygen and then as innocuous Components are released into the environment.

to Exploitation of the oxygen storage effect, it is necessary to change the mixture from a lean to a fat area and vice versa to ensure around the catalyst alternating with oxygen and reducing agents to act on. In the prior art it is customary for mixture control breakthroughs of oxygen or reducing agents that then occur if the storage capacity of the catalyst for Oxygen exhausted or the oxygen stored in the catalyst is used up. To detect the fat or lean breakthroughs downstream of the catalyst is usually an oxygen probe arranged. Due to the principle occur in this Process downstream of the catalyst unconverted exhaust masses either fired into the environment or by another Catalyst must be cleaned.

The Grease or lean exhaust gas breakthroughs occur because at a time of the mixture change still a certain amount of lean or rich exhaust between the engine outlet and the oxygen sensor in the exhaust system is located. For example, if the engine is switched to slightly rich operation is, this has no effect on a lean amount of residual gas in the catalyst and in the exhaust system.

From the DE 101 25 759 A1 It is known to determine the loading state of a NOx storage catalytic converter, taking into account a NOx discharge from the NOx storage catalytic converter. In this case, the NOx discharge is determined as a function of a measured or modeled concentration of reducing agents in the exhaust gas present downstream of the NOx storage catalytic converter.

It is from the DE 44 42 734 C2 a control system for an internal combustion engine is known, in which an amount of storage of oxygen stored in the catalyst is estimated. A physical amount correlated with the purification degree of the catalyst is calculated on the basis of the estimated oxygen storage amount, and based on the calculated physical amount, an air-fuel ratio to be supplied to the engine is determined. Furthermore, from the DE 196 06 652 A1 a method for adjusting the air / fuel ratio for an internal combustion engine with downstream, capable of oxygen storage catalyst known. The oxygen components in the exhaust gas before and after the catalyst are detected and a measure of the instantaneous oxygen filling level of the catalyst is modeled from said oxygen fractions. Statements about the aging state of the catalyst are derived from the model parameters and the fuel air ratio is set so that the oxygen content of the catalyst is kept at a constant medium level.

From the DE 102 25 937 A1 a method is known in which a desired oxygen loading of a catalyst is to be achieved in order to avoid pollutant breakthroughs even in the event of deviations from a desired lambda value.

task It is the object of the present invention to optimally utilize the Oxygen storage capacity a arranged in the exhaust system of an internal combustion engine catalyst with a simultaneous reduction of pollutant emissions downstream to reach the catalyst.

The The object is achieved by the Characteristics of the independent Claims solved.

According to the invention is a Mixture change depending one from a model of oxygen storage capability calculated value of the oxygen loading initiated. The mixture change unlike in the prior art is not involved in the determination of breakthroughs bound by exhaust gases on the catalyst. Therefore, another one is Catalyst downstream of the first catalyst is not essential to ensure sufficient overall conversion. This can result in significant cost reductions and savings are achieved in particular on precious metals for additional catalysts.

In addition, when a mixture change is initiated by means of an oxygen sensor arranged downstream from the catalyst in a rich or lean breakdown, a higher emission reliability in the case of deviations between the oxygen storage type according to the model expected value of the oxygen loading and the actual value of the oxygen loading can be achieved.

In a preferred embodiment the model of oxygen storage capacity is adapted to changes the oxygen storage capacity with to take into account the time. If an adapted value of the oxygen storage capacity dynamically dependent changed by optimization parameters is, can be misadaptations avoided.

The The invention further comprises a device for mixture control for execution the method according to the invention.

in the Following are further advantages and aspects of the invention as well independently from its summary in the claims with reference to drawings shown.

It demonstrate

1 a catalyst system according to the prior art

2 a control function of a catalyst system according to the prior art

3 characteristic waveforms of oxygen probes upstream and downstream of a prior art catalyst

4 a catalyst system according to the invention

5 Control functions according to the invention

6 a waveform according to the invention

7 an adaptation curve for a value of the maximum oxygen storage capacity

8th an optimized adaptation strategy for a maximum oxygen storage capacity value.

In the following, in contrast to a lambda control according to the prior art ( 1 to 3 ) presented the method according to the invention, which makes it possible to better utilize the existing catalyst coating with reduced expenditure of lambda probes. Thus, there is the possibility to reduce precious metal costs, to reduce the design to just one converter, to reduce the cat volumes or to use less lambda probes.

In 1 is a catalyst system of an internal combustion engine 1 represented according to the prior art. Downstream of the engine 1 is a first catalyst 2 arranged with a lambda probe arranged upstream of the catalytic converter 3 , Downstream of the catalyst 2 is another catalyst 2a and one between the catalysts 2 and 2a positioned lambda probe 3a intended. The catalyst 2a Downstream is a lambda probe 3b , The lambda probes 3 . 3a and 3b give signals to an engine control unit 4 from. Between the engine 1 and the engine control unit 4 become signals 5 and 6 replaced. The signals 5 relate to an exhaust gas mass flow, an air mass, a fuel mass, signals representing the catalyst and engine temperature, an exhaust gas recirculation rate, the position of any existing charge movement flap, engine speed, driving speed, camshaft adjustment or the like. signals 6 serve the mixture precontrol, in a direct injection gasoline engine, for example, an injection time.

In 2 are essential functions of the engine control unit 4 through module blocks 4a . 4b illustrated. 4a denotes a mixture precontrol, 4b an adjustment function for the mean value of the mixture. Using signals from the lambda probe 3a in the module 4b is calculated, as shown in more detail below, a mixture change W from lean to rich or vice versa. module 4b receives from Modul 4a at least part of the signals 5 that the module 4a from the engine 1 to provide. module 4a receives information about a lambda lambda value from the Lambda probe 3 upstream of the catalyst 2 , The motor 1 is operated in such a way that it alternately emits slightly lean or slightly rich exhaust gas, see signal 3 in 3 with typical values for λ = 1.02 to 1.05 and λ = 0.98 to 0.95. This results in typical values for amplitude a in 1 from 0.02 to 0.05.

In 3 are characteristic waveforms of the probes 3 and 3a according to the system in 1 shown. The upper part of the figure shows the course of the lambda value upstream of the catalytic converter. Here, p denotes a period length of a fat-lean interval, a an amplitude of the lambda value into the fat or lean, v a delay time or phase shift of the mixture downstream of the catalyst 2 and s is a system time constant that describes mixing and damping of the exhaust gas as it passes through the catalyst system.

The changeover W from lean to rich (time t1) or vice versa (time t2) is determined by the engine control functions 4a always triggered when the probe 3a the breakthrough of a Component (lean at time t1, or fat at time t2) detects. The detection of the breakthrough is usually associated with a particular probe voltage, which emits a usually built λ jump probe. The engine control functions 4a give either immediately or after a freely determinable waiting time ( 4b ) change the momentum of the mixture from lean to rich (t1) or vice versa. The delay of the pulse for switching can be used if the mixture is to be trimmed on average to a somewhat richer, or somewhat lean value. This adjustment may possibly also via the probe 3b be effected. The switchover between rich and lean is carried out by the mixture precontrol, for example via the variation of the injected fuel quantity ( 4a and 6 ).

The mixture precontrol 4a The lean or rich mixture can be either via a lambda control using the probe 3 control (closed loop) or even without feedback (open loop). For the latter case, it would be possible to use the probe 3 with sacrifices in the accuracy of the adjustability of the mixture before the catalyst 2 ,

The signal of the lambda probe 3a is opposite the signal 3 changed. This is due to the route being the catalyst 2 , in particular including its storage effect for oxygen, maps. The signal 3a is opposite the signal 3 shifted by the duration v, which reflects the gas transit time through the catalyst (phase shift / dead time). Gas mixing operations also attenuate the signal, which is represented by the system time constant s. In addition, a smaller amplitude a2 is measured than before the catalyst 2 in which the conversion properties of the catalyst are reflected. The period length p is of the size of the oxygen storage of catalyst 2 dependent.

The parameters p, s, a2 and possibly also v change over the life of the catalyst 2 if it is damaged or decreases in its conversion efficiency due to aging. Also variations of temperature, mixture composition and exhaust mass flow of the engine have an influence on the mentioned parameters.

The in 1 illustrated catalyst 2a and the probe 3b are not directly integrated in the primary mixture control circuit. The catalyst 2a has for the illustrated control the task to convert the principle occurring occurring fat, or lean gas breakthroughs. For example, lean gas breakthroughs occur because, at the time of the mixture change (t1), there is still a certain amount of lean exhaust gas between the engine outlet and the probe 3a located in the exhaust system. The engine is switched to slightly rich operation, but no longer has any influence on the mentioned lean residual gas in the cat and in the pipes. This quantity is ejected without conversion, in particular with regard to NOx, or must be in the downstream catalytic converter 2a be converted. However, this is only effective once it has reached its operating temperature. About the probe 3b is usually the function of the second catalyst 2a diagnosed and / or operated a higher-level control loop for more accurate determination of the catalyst optimum. From information of the higher-order control loop, the waiting time described above ( 4b ) determines which mixture control is to wait before changing between rich and lean or changing the assigned probe voltage at which a mixture breakthrough with probe 3a is recognized.

In 4 is a schematic representation of a preferred construction according to the invention shown with only one lambda probe 3a and a catalyst 2 as necessary components. The lambda probes 3 . 3b and the downstream of the catalyst 2 arranged catalyst 2a are optional.

As in the prior art takes place between the engine 1 and the engine control unit 4 a signal exchange of operating parameters such as exhaust gas mass flow, air mass, fuel mass, catalyst and engine temperature, exhaust gas recirculation rate, position of the charge motion flap, engine speed, driving speed, camshaft adjustment, etc. A module 4a sets over signals 6 For example, the injection time, which may be different in general between lean and rich, fixed.

In 5 are control functions of the engine control unit 4 represented according to the invention. A mixture precontrol 4a gets from the engine 1 the mentioned signals 5 ,

Further, a control module 4c provided in which a model of the oxygen storage capacity of the catalyst is implemented. Further, a catalyst controller 4d provided the signals from the lambda probe 3a receives. The modules 4c and 4d get at least a part of the mentioned motor signals 5 , in the 5 when 5a designated. In particular, receives module 4c Data on the input lambda values of the catalyst as well as further catalyst parameters for the determination of the oxygen loading (filling level F) of the catalyst and for the adaptation of the oxygen storage capacity model. The input lambda values can be pilot lambda values or - if a probe 3 is present - be measured values of the exhaust gas.

The module 4d the level F determined by the oxygen storage capability model is transmitted. In module 4d a signal W is generated, which, depending on the calculated level F, an instruction for a mixture change between rich and lean mixture to the module 4a transmitted.

Further, in the module 4d generates an adaptation information A with which the model of the oxygen storage capability is adapted.

In 6 a waveform is shown by the method according to the invention. In the upper part of the figure, the course of the lambda value is upstream of the catalyst 2 shown. At time t ₁ , a mixture change takes place. The time t ₁ is chosen so that no major pollutant breakthrough on the catalyst 2 occurs. The signal of the lambda probe 3a is in the lower part of the 6 shown. As in prior art driving, there is a gas running time V between the signals 3 and 3a to observe. The switching time W is earlier by an amount x compared to the prior art method, since breakdown of exhaust gas components is minimized.

at the model of the oxygen storage capacity of the catalyst in The engine control is assumed to be of a maximum value but over the life, at different temperatures or due other influences changes. Therefore have to certain model parameters are always adapted (adapted).

• 1. Gemischwechsel durch Modell initiiert: Bei einem vordefinierten Füllstand F zwischen 0 und 100% des möglichen Gesamt-(Sauerstoff-)speichers gibt das Modul 4d ein Signal zum Gemischwechsel W an die Motorsteuerungsfunktionen weiter. Mögliche Füllstände für einen vom Modell initiierten Wechsel wären z. B. 5% und 95% OSC. In einem mageren Motorbetrieb wird Sauerstoff in das Modell eingetragen und der Füllstand steigt. Bei 95% wird der Gemischwechsel ins Fette angestoßen. Das zu diesem Zeitpunkt in Rohr und Katalysator verbleibende magere Abgas füllt den Speicher noch auf annähernd 100%, so dass stromab des Katalysators 2 an der Sonde 3a kein so großer Magerdurchbruch (wie im Stand der Technik) erkannt wird. Durch das jetzt fette Abgas werden die Sauerstoffkomponenten im Katalysator reduziert, bis der berechnete Füllstand bei 5% liegt und durch das Modell wider der Gemischwechsel angestoßen wird, etc. Zusätzlich kann noch eine gewisse Sicherheit gegen Durchbrüche durch Ungenauigkeiten von Messung und Vorsteuerung berücksichtigt werden. Der Füllstand, bei dem der Wechsel angestoßen wird, kann entweder festgeschrieben sein (z. B. 5% und 95%) oder von verschiedenen Betriebsparametern 5a abhängen. Dies hat den Hintergrund, dass z. B. im stationären Fahrbetrieb die Gemischvorsteuerung sehr genau arbeitet. In diesem Fall kann der Katalysatorspeicher voll ausgenutzt werden. Im dynamischen Fahrbetrieb arbeitet die Gemischvorsteuerung i. A. ungenauer und die berechneten Fett- und Magergasmengen können von den real in den Katalysator eingetragenen abweichen. Für diese Fälle würde erfindungsgemäß der Gemischwechsel früher angestoßen (z. B. bei 30% und 70%), um Gemischdurchbrüche durch Vorsteuerfehler zu vermeiden. Das Modell kann vorzugsweise Informationen über Alterungseffekte des Katalysators, insbesondere der Sauerstoffspeicherabnahme, aus einer externen Katalysatordiagnose erhalten, aus vorab definierten Kennlinien und Kennfeldern auslesen oder aber wie im Folgenden beschrieben selbst bestimmen.
• 2. Vorzeitiger Durchbruch von Sonde 3a erkannt: Sollte entgegen der Berechnung das Modell ein Fett- oder Magergasdurchbruch durch den Katalysator von Sonde 3a erkannt werden, dann wird ein Gemischwechsel durch das Signal der Sonde 3a initiiert und nicht erst durch den berechneten Füllstand des Modells. Dieser Zustand kann durch Fehler in der Gemischvorsteuerung oder auch durch eine fehlerhafte Aufkumulierung im Modell entstehen. Erfindungsgemäß wird das Modell wieder angepasst, d. h. der berechnete Füllstand mit den realen 100% zum Zeitpunkt des Durchbruchs. Gleichzeitig kann auch die maximal mögliche Gesamtspeichergröße angepasst werden, indem eine Adaptionsinformation A an das Modell gegeben wird.
• 3. Geplante Adaption: Nach einer gewissen Fahrzeug-Laufzeit, nach einer bestimmten Anzahl an Mager-Fett Zyklen, nach einer bestimmten durchgesetzten Luftmasse (Schwellwert SW_P) oder einer bestimmten Anzahl unvorhergesehener Durchbrüche (Schwellwert SW_D), wie in 2. beschrieben, wird das Modell adaptiert. Dann wird nicht, wie in 1. dargestellt, der Gemischwechsel vor einem gemessenen Durchbruch durch Modell und Algorithmus angestoßen, sondern es wird aktiv gewartet, bis die Sonde 3a einen größeren Durchbruch misst. Aus der gewonnenen Information über die Dauer bzw. die Abgasmasse, werden Parameter, insbesondere die maximale Größe des Speichers, angepasst.

According to the invention, the following three possible states for the mixture change result:

• 1. Mixture change initiated by model: At a predefined level F between 0 and 100% of the possible total (oxygen) storage, the module gives 4d a signal to the mixture change W to the engine control functions on. Possible levels for a change initiated by the model would be z. 5% and 95% OSC. In a lean engine operation, oxygen is added to the model and the level rises. At 95% the mixture change into the fat is triggered. The lean exhaust gas remaining in the pipe and catalyst at this time fills the reservoir to approximately 100%, so that downstream of the catalyst 2 at the probe 3a no such large lean breakdown (as in the prior art) is detected. The now rich exhaust gas reduces the oxygen components in the catalytic converter until the calculated level is 5% and is triggered by the model against the mixture change, etc. In addition, some safety against breakthroughs due to inaccuracies in the measurement and pilot control can be taken into account. The level at which the change is initiated can be either fixed (eg 5% and 95%) or different operating parameters 5a depend. This has the background that z. B. in stationary driving the mixture feedforward works very accurately. In this case, the catalyst storage can be fully utilized. In dynamic driving mode, the mixture precontrol i works. A. inaccurate and the calculated amounts of fat and lean gas may differ from the real registered in the catalyst. For these cases, according to the invention, the mixture change would be triggered earlier (eg at 30% and 70%) in order to avoid mixture breakthroughs due to pilot control errors. The model may preferably obtain information about aging effects of the catalyst, in particular the oxygen storage decrease, from an external catalyst diagnosis, read from previously defined characteristic curves and maps or else determine it as described below.
• 2. Premature breakthrough of probe 3a recognized: Should contrary to the calculation the model a fat or lean gas breakthrough by the catalyst of probe 3a be detected, then a mixture change by the signal of the probe 3a initiated and not only by the calculated level of the model. This condition can be caused by errors in the mixture precontrol or by an incorrect accumulation in the model. According to the invention, the model is adjusted again, ie the calculated level with the real 100% at the time of breakthrough. At the same time, the maximum possible total memory size can also be adapted by giving adaptation information A to the model.
• 3. Planned adaptation: After a certain vehicle running time, after a certain number of lean-rich cycles, after a certain enforced air mass (threshold SW_P) or a certain number of unforeseen breakthroughs (threshold SW_D), as described in 2., the Model adapted. Then, as shown in FIG. 1, the mixture change is not initiated prior to a measured breakthrough by the model and algorithm, but it is actively maintained until the probe 3a measures a bigger breakthrough. From the information obtained over the duration or the exhaust gas mass, parameters, in particular the maximum size of the memory, adjusted.

In 7 is a possible adaptation curve of the size of the (oxygen) total storage over the vehicle life shown. Normally, the adapted value decreases with each adaptation process, as well as the real oxygen memory capability continues to be reduced. However, if erroneous adaptation processes are triggered by mixture breakthroughs (dotted line), it takes a very long time until the adapted value again corresponds to the real value of the oxygen storage. Parts of the memory thus remain unused for a long time, in 7 represented by dotted area. Faulty adaptation processes can be caused by a mixture breakthrough, which was caused by a faulty mixture precontrol or other inaccuracies.

Therefore, in a preferred embodiment, the adaptation value is repeatedly increased as a function of the time of an enforced air mass, number of mixture changes without breakthrough or similar parameters, and thus prematurely triggers an adaptation. In 8th a corresponding adaptation strategy is shown. The adaptation value increases between the individual adaptation processes. Faulty adaptation values are thus quickly adapted again in real conditions and the oxygen storage is used better over life as the comparison of the size of the dotted areas in 7 and 8th shows.

Claims (13)

A method for mixture control in an internal combustion engine with at least one disposed in an exhaust system of the internal combustion engine having an oxygen storage capability catalyst and an oxygen sensor arranged downstream of the catalyst, wherein a model of oxygen storage capability is provided, which in dependence of input lambda values and catalyst parameter values a value of Calculated oxygen loading of the catalyst, characterized in that a mixture change in dependence (a) of the calculated value of the oxygen loading and (b) of a determined by means of the oxygen sensor fat or lean breakdown at the catalyst is initiated. Method according to claim 1, characterized in that that adaptation information for adaptation of the model using from at least one of the following information - Results a catalyst diagnosis - Evaluation results of lean or fat breaks on the catalyst is determined. Method according to claim 1 or 2, characterized that an adaptation of the model occurs when at least one of meets the following conditions is: - Number the determined fat or lean breakthroughs on the catalyst is greater than a threshold SW_D, - one Value of at least one of the following operating parameters, vehicle running time, Lean fat cycles, air mass flow rate is greater than a threshold SW_P. Method according to at least one of the preceding Claims, characterized in that in the case of a determined fat or Lean breakdown at the catalyst the value of calculated oxygen loading is adjusted using evaluation results of the breakthrough. Method according to one of claims 3 to 4, characterized that a time duration and / or an exhaust mass of the lean or Fat breakthrough determined and included in the evaluation results becomes. Method according to at least one of the preceding Claims, characterized in that for carrying out a planned adaptation the mixture change for a given time interval not dependent on the calculated Value of oxygen loading according to claim 1 (a) is initiated. Method according to at least one of claims 2 to 6, characterized in that an adapted value of the oxygen storage capacity dynamically dependent changed by optimization parameters becomes. Method according to claim 7, characterized in that that the adapted value of the oxygen storage capacity in dependence increased by optimization parameters becomes. Method according to at least one of the preceding Claims, characterized in that input lambda values from a pilot control device or an oxygen probe arranged upstream of the catalyst to be delivered. An apparatus for mixture control in an internal combustion engine having an arranged in an exhaust system of the internal combustion engine, an oxygen storage capability catalyst and an oxygen sensor arranged downstream of the catalyst, wherein a model of the oxygen storage capability is provided, which in dependence on input lambda values and catalyst parameter values a value of an oxygen loading of the catalyst, characterized in that control modules are provided, which in a mixture change depending on - the calculated value of the oxygen loading or - can initiate a fat or lean breakdown on the catalyst determined by means of the oxygen sensor. Device according to claim 10, characterized in that that the catalyst is a 3-way catalyst. Device according to claim 10 or 11, characterized that downstream of the catalyst arranged another catalyst is. Device according to one of claims 10 to 12, characterized that the internal combustion engine is a direct-injection, preferably lean-running or layer loadable Otto engine is.