DE102010033335B4 - Method for determining the oxygen storage capacity of an oxygen storage device assigned to a catalyst - Google Patents

Abstract

Method for determining the oxygen storage capacity of an oxygen storage device (4) assigned to a catalytic converter (3) in an arrangement in which, in the outflow direction of the exhaust gas of an internal combustion engine (1), a pre-catalytic converter (5) in front of the catalytic converter (3) and behind at least a section of the catalytic converter ( 3) a post-catalytic converter (6) measures the air-fuel ratio, in which case the oxygen storage (4) is initially charged with rich exhaust gas under control using the pre-catalytic converter (5) in order to empty it of oxygen as much as possible , and then there is a change to a loading of the oxygen storage with lean exhaust gas in order to fill it again with oxygen, the oxygen quantity entered per time interval being integrated over a time interval which begins with the time (to) of the change and with a time (t3) ends at which the oxygen reservoir is filled when it falls below an S. threshold value in an output signal of the post-cat lambda probe (6), characterized in that the integral obtained in this way is corrected, with a time offset (t1-to) between the point in time of the change and a change in the sign of the slope of the output signal of the post-cat probe ( 6) is measured and taken into account when making corrections.

Description

The invention relates to a method for determining the oxygen storage capacity according to the preamble of claim 1.

The starting point is therefore the situation that a catalytic converter is arranged downstream of an internal combustion engine in the outflow direction of the exhaust gas, an oxygen storage device being assigned to the catalytic converter; the oxygen reservoir can in particular be integrated into the catalytic converter, but can also be provided as a separate component. A so-called pre-catalyst lambda probe is located in front of the catalytic converter, and a post-catalyst lambda probe is located behind the catalyst. Lambda sensors measure the air-fuel ratio (in the exhaust gas).

A method for determining the oxygen storage capacity is described below. To determine the oxygen storage capacity, all that needs to be done is to replace “rich” with “lean” and vice versa, and whenever oxygen is introduced, the procedure for determining the oxygen storage capacity is based on the assumption that oxygen will be discharged.

To determine the oxygen (storage) storage capacity, the oxygen storage tank is initially charged with rich exhaust gas in order to empty it of oxygen. The fact that the exhaust gas is rich is determined on the basis of signals from the pre-catalyst lambda probe, i.e. under control on the basis of the same.

When the oxygen storage tank is almost completely emptied (according to a predetermined criterion), there is an immediate change to loading the oxygen storage tank with lean exhaust gas in order to gradually fill it with oxygen again. Here too, the “lean” air-fuel ratio is set under control using the pre-catalytic converter.

The amount of oxygen entered per time interval during refilling is then integrated. If possible, the entire amount of oxygen should be recorded. Accordingly, the time interval for the integration begins with the point in time of the change (which is determined by a controller and is therefore known). The time interval ends at a point in time at which the fact that the oxygen storage tank is full has an effect on the output signal of the post-cat probe falling below a threshold value. As long as the oxygen reservoir is not yet completely filled, the oxygen is taken from the lean exhaust gas by the oxygen reservoir from the exhaust gas, and the exhaust gas is no longer lean when it reaches the post-cat sensor. However, if the oxygen storage tank is filled at some point, then no more oxygen is stored and lean exhaust gas actually arrives at the post-cat probe. In the voltage signal, z. B. a voltage of typically 0.4 V falls below.

The method according to the invention works extremely well as long as the measuring devices used are fully functional.

The problem is that the pre-catlambda probe is not fully functional EP 05 96 635 B1 , German as DE 693 27 148 T2 released. Aging of the pre-catlambda probe is compensated for by deliberately delaying a signal.

The DE 10 2004 009 615 B4 describes a method for determining the current oxygen loading of a 3-way catalytic converter. This method takes into account that the lambda value can be subject to measurement errors. In particular, the method also includes an integration over time.

If the post-cat lambda probe has aged, it can react with a delay in particular. If the start of the time interval is the point in time of the change, then the measurement starts at the correct point in time, but ends with a delay due to the aging of the post-catlambda probe, because the point in time at which the interval ends is determined by signals from this post-catlambda probe.

The DE 10 2004 062 408 A1 describes a method for determining an oxygen storage capacity of an exhaust gas catalytic converter, in which a predetermined first rich air-fuel ratio is first set in a combustion chamber of a cylinder of an internal combustion engine, and a first oxygen storage capacity value is determined. In addition, a predetermined second rich air-fuel ratio is set in the combustion chamber of the cylinder later, and a second oxygen storage capacity value is determined. A corrected oxygen storage capacity value is then determined using the two oxygen storage capacity values. The DE 10 2004 062 408 A1 discloses that a dynamic time period can also be calculated using the two oxygen storage capacity values, which corresponds to the delay time in the response time periods of two exhaust gas probes. A formula for this delay time contains the difference between the two determined oxygen storage capacity values.

The US patent application US 2009/0182490 A1 describes that as a measure of the poor functioning of a lambda probe, the time period between a brought about change in the air-fuel ratio and a point in time at which the output signal of the lambda probe has a turning point.

The DE 10 2008 023 893 A1 describes a method for diagnosing the functionality of a step probe, which describes the knowledge that when there is a change from a lean to a rich exhaust gas mixture in a small time interval near the lambda jump of a pre-catalyst lambda probe in the signal profile measured by a post-catalyst lambda probe , a characteristic behavior is visible. The functionality of the lambda probe can be deduced from the signals at small intervals around the changeover time, without having to consider the actual probe jump in the curves.

Furthermore, the US 2009/0235726 A1 a method for determining the oxygen storage capacity of a catalytic converter for a motor vehicle and an associated measuring device.

The object of the invention is to show a way in which aging of the post-cat lambda probe can be explicitly taken into account when determining the oxygen storage capacity.

The object is achieved by a method with the features according to patent claim 1.

The integral obtained is correspondingly corrected according to the invention. To correct this, a time offset is first measured between the point in time of the change and a change in the sign of the slope of the output signal of the post-cat probe. This time offset is then taken into account when making corrections.

Here one makes use of the knowledge that the change in the exposure - clearly recognizable at the end of the time interval - can already be determined relatively directly to a sufficient extent on the output signal of the post-cat lambda probe. Ideally, immediately with the change in the application, there is a change in the sign of the slope (first time derivative) of the output signal. However, if such a change occurs with a time delay, this indicates that the probe is aging.

The time delay ("probe delay") remains constant over the entire time. Therefore, based on the time offset at the time of the change, you can also determine when the measurement should have ended. The measurement ends too late at exactly the time offset.

In the event that a large amount of storage capacity can be made available (e.g. in a ring memory), intermediate values of the integral can be retained during the integration over time, and the time offset to achieve a final value can then be deducted from the point in time at the end of the time interval will. The intermediate value assigned to the final value can be used as a correction value. With this variant, one cancels the correct value for the integral, so to speak.

However, sufficient storage capacity is not always available. If you want to save the weight for a ring memory or another high-capacity data memory, you cannot keep all intermediate values of the integral and then you have to deduce the value that was calculated at the “right time”. The proportion of calculated oxygen storage capacity which is included in the integral over a duration equal to the length of the time offset before the end of the time interval is therefore preferably calculated. Exactly this portion is then subtracted from the integral to achieve a correction value.

In a simple case, the oxygen loading of the oxygen reservoir increases continuously, then the correction value can be obtained by simple proportional calculation based on the time relationships.

However, it can now be the case that the method according to the invention should be carried out during any journey. It is then possible that the vehicle driver changes the exhaust gas mass flow during the measurement or that the air-fuel ratio changes due to a change in the load speed point. In this case, the assumption that the slope in the oxygen charge remains constant is no longer correct.

A change in the exhaust gas mass or in the air-fuel ratio that occurred during the time interval is preferably taken into account when estimating the proportion.

The effect of a time delay becomes noticeable in a low-pass filter. In particular, the time offset can predetermine or in particular be a filter constant for such a (digital) low-pass filter. If the parameter “oxygen entry per time” is filtered with this digital low-pass filter, a result of the filtering is obtained at the end of this time interval that can be used as the slope of a straight line. If you multiply this slope by the time offset, you get the proportion of the calculated oxygen storage capacity that is excess.

• 1 eine Anordnung zeigt, bei der ein Einsatz des erfindungsgemäßen Verfahrens sinnvoll ist,
• 2A das Luft-Kraftstoff-Verhältnis Lambda zeigt, wie es beim erfindungsgemäßen Verfahren festgelegt ist,
• 2B Antwortsignale einer voll funktionsfähigen und einer mit Zeitverzug behafteten Nachkatlambdasonde auf die Beaufschlagung gemäß 2A,
• 2C die Sauerstoffbeladung, wie sie bei Gegebensein dieser beiden Antwortsignale jeweils berechnet werden würde,
• 3A und 3B den 2A und 2B entsprechende Darstellungen sind,
• 3D ein beispielhafter Abgasmassenstrom ist, wie er durch den Fahrzeugführer eingestellt werden kann, und
• 3C eine der 2C für den Fall der 3D entsprechende Darstellung ist,
• 4A und 4B den 2A und 2B entsprechende Darstellungen sind,
• 4D ein Abgasmassenstrom in einem weiteren Beispiel ist, wie er durch den Fahrzeugführer eingestellt werden kann, und
• 4C eine der 2C für den Fall der 4D entsprechende Darstellung ist,
• 5A und 5B den 2A und 2B entsprechende Darstellungen sind,
• 5D eine der 4D entsprechende Darstellung ist,
• 5E der einer digitalen Tiefpassfilterung unterzogene Sauerstoffeintrag ist, und
• 5C für den Fall der 5D und 5E eine den 2C, 3C und 4C entsprechende Darstellung ist.

The invention is described in accordance with preferred embodiments with reference to the drawing in which

• 1 shows an arrangement in which use of the method according to the invention is useful,
• 2A shows the air-fuel ratio lambda as it is determined in the method according to the invention,
• 2 B Response signals of a fully functional post-catlambda probe and a post-catalytic converter with a time delay to the application according to 2A ,
• 2C the oxygen load as it would be calculated if these two response signals were given,
• 3A and 3B the 2A and 2 B corresponding representations are,
• 3D is an exemplary exhaust gas mass flow as it can be set by the vehicle driver, and
• 3C one of the 2C in case of 3D corresponding representation is
• 4A and 4B the 2A and 2 B corresponding representations are,
• 4D is an exhaust gas mass flow in a further example, as it can be set by the vehicle driver, and
• 4C one of the 2C in case of 4D corresponding representation is
• 5A and 5B the 2A and 2 B corresponding representations are,
• 5D one of the 4D corresponding representation is
• 5E is the oxygen input subjected to digital low-pass filtering, and
• 5C in case of 5D and 5E one the 2C , 3C and 4C corresponding representation is.

1 shows a schematic representation of an internal combustion engine 1 with an exhaust line 2 . The exhaust system 2 includes a catalytic converter 3 , the z. B. is designed as a three-way catalyst, as a NOx storage catalyst or as an active particle filter, as well as an integrated oxygen storage 4th includes. The exhaust system 2 further comprises an upstream of the catalytic converter 3 arranged pre-catalytic converter, which serves as a guide probe, and one for the catalytic converter 3 associated post-catlambda probe 6 that serves as a control probe.

The post-katlambda probe 6 is in the present exemplary embodiment downstream of the catalytic converter 3 arranged. This post-catalytic converter could just as well be installed directly in the catalytic converter 3 , ie according to a partial volume of the oxygen storage 4th , be arranged.

It is assumed below that the exhaust gas from the internal combustion engine 1 can be adjusted to a predetermined air-fuel ratio lambda at least with a predetermined accuracy.

It should increase the oxygen storage capacity of the oxygen storage system 4th be determined.

The oxygen storage capacity can be determined while oxygen is being stored or while it is being withdrawn. Determination of the oxygen storage capacity during the storage is described below.

So that the oxygen storage capacity can be measured during the storage, the oxygen storage 4th first be completely emptied. For this purpose the internal combustion engine 1 operated so that that in the catalyst 3 exhaust gas entering is rich, and a corresponding curve 10 is in 2A shown: There is initially an excess of fuel in the exhaust gas. The excess fuel burns, removing oxygen from the oxygen reservoir, which is thus gradually emptied.

At some point, the oxygen reservoir is almost completely empty, and it is possible to switch to the application of lean exhaust gas, i.e. with exhaust gas in which there is an excess of air in relation to the fuel and thus also an excess of oxygen. The change from fat to lean takes place at time t ₀ . Such a change has an effect on the signal of the post-catlambda probe 6 according to curve 12 to the effect that exactly at time t _{0 there is} a change in the sign of the time derivative. Is the post-Katlambda probe 6 not fully functional, but reacts with a time delay ("probe delay"). B. the curve 14th , according to which the change in the sign of the time derivative only takes place at a point in time t ₁ after the point in time t ₀ .

The oxygen storage capacity is to be determined during the storage of oxygen, that is to say from time t ₀ . In the present case, it is initially a prerequisite that the exhaust gas mass flow is kept constant. Since the air-fuel ratio lambda is kept constant, a constant amount of oxygen is entered per time interval. The oxygen loading OSC thus increases monotonically with time, see the curve 16 .

wobei über die gesamte Zeit der Sauerstoffeintrag von t₀ bis t_End integriert wird. Wenn λ = cst. und ṁ = est., steigt OSC linear mit dem Integralende t_End, wie anhand der Kurve 16 auch ersichtlich.The oxygen storage capacity OSC is generally calculated using the following formula: $$O\mathrm{S.}\mathrm{C.}={\displaystyle \underset{t_0}{\overset{t_{\mathrm{E.}nd}}{\int }}0.23\left(1-\lambda \left(t\right)\right)dt*\dot{m}\left(t\right)dt,}$$

the oxygen input from t ₀ to t _{end being} integrated over the entire time. If λ = cst. and ṁ = est., OSC increases linearly with the _{end of} the integral t _End , as shown by the curve 16 also evident.

It is now questionable when the integration should end.

Conventionally, the integration ends when the post-cat probe measures a voltage that is less than a certain threshold value, e.g. B. as 0.4 V. In this case, namely, the lean exhaust gas goes directly to the post-cat sensor without releasing any further oxygen, which means that the oxygen storage tank is completely full.

In the case of a fully functional probe, time t _{2 is} exactly the right time. If the probe is not fully functional, the voltage falls below 0.4 V too late, namely only at time t ₃ .

If the time derivative is integrated from the point in time of the change, the integral in the case of a constant lambda and exhaust gas mass flow is independent of whether the post-catalytic converter is used 6 is fully functional or not: the integral t ₀ to t ₂ is exactly the same as the integral from t ₁ to t ₃ .

The situation is different if the lambda value and the exhaust gas mass flow change during the time interval: using the 3A to 3D it can be seen that with a change in the exhaust gas mass flow ṁ according to the curve 18th in the integration according to the present formula, there is no longer a constant slope: between time t ₀ and t ₁ , the slope is greater in the present example than subsequently between t ₁ and t ₂ or t ₃ . The "correct" value for the oxygen storage capacity would be at the point 20th measured. Are you dealing with an aged, not fully functional post-Katlambda probe 6 to do, however, one measures an oxygen storage capacity according to the point 22nd .

The oxygen storage capacity is measured too low if the exhaust gas mass flow decreases in the meantime.

Is accordingly 4A to 4D it can be seen that when the catalyst is subjected to an exhaust gas mass flow according to the curve 24 the "correct" oxygen storage capacity as at the point 26th would be measured is lower than what is actually measured at the point 28 .

As a solution to this problem, it is proposed in the present case to have the measurement of the oxygen storage capacity of the oxygen storage capacity basically begin at time t _{0 of} the change, for which purpose the signal from the post-catalytic converter is not used 6 can use, but the signal of the precatlambda probe 5 must use.

In this case, if the post-catalytic converter probe is aged, an excessively high oxygen storage capacity is measured, since the integral is calculated each time up to time t ₃ . It must therefore be considered in some way what amount of oxygen is entered between the times t ₂ and t ₃ .

In a simple embodiment, each intermediate value for the integral OSC is simply stored. Due to the change in the sign of the signal from the post-Katlambda probe according to the curve 14th the time t ₁ can be determined, thus also the distance t ₁ - t ₀ . The time t ₃ is also known, and the time t ₂ = t ₃ - t ₁ + t _{0 can then be} calculated according to t ₃ - t ₂ = t ₁ - t ₀ .

If the time t _{2 is} known, the actual value for the oxygen storage capacity can be inferred. In the simplest case, all intermediate values are simply stored during the integration of the variable OSC, that is to say, starting from the point in time t ₀ for several times t _i in discrete intervals that are relatively small in relation to the total time. If these intermediate values are stored, it is still possible to say at time t ₃ what value the integral had at time t ₂ , and then the correct value for the oxygen storage capacity has been obtained.

You can't always keep such a large amount of data in stock. It is therefore preferably provided that the integral is calculated up to the point in time t ₃ and is then calculated back to determine how large it would have been at the point in time t ₂ . In the case of the curve 24 a digital low-pass filtering can be carried out in which the filter constant is determined by t ₁ - t ₀ . If you then filter the amount of oxygen introduced per time using the low-pass filter, you get the curve 30th . When calculating the oxygen storage capacity OSC one arrives at a point 32 , and it can then be based on the value 34 out of the curve 30th , determines a gradient of a route at time t ₃ 36 between the point 32 and a point to be calculated 38 be determined.

If the slope is multiplied by t ₃ - t ₂ , i.e. t ₁ - t ₀ , a value ΔOSC is obtained between the point 32 and the point 38 . If one has calculated the oxygen storage capacity OSC from t ₀ to t ₃ and then subtracts the quantity ΔOSC, one arrives at the point 38 and knows the actual oxygen storage capacity.

Claims (7)

Method for determining the oxygen storage capacity of an oxygen storage device (4) assigned to a catalytic converter (3) in an arrangement in which, in the outflow direction of the exhaust gas of an internal combustion engine (1), a pre-catalytic converter (5) in front of the catalytic converter (3) and behind at least a section of the catalytic converter ( 3) a post-catalytic converter (6) measures the air-fuel ratio, in which case the oxygen storage (4) is initially charged with rich exhaust gas under control using the pre-catalytic converter (5) in order to empty it of oxygen as much as possible , and then there is a change to a loading of the oxygen storage with lean exhaust gas in order to fill it again with oxygen, the oxygen quantity entered per time interval being integrated over a time interval which begins with the time (to) of the change and with a time (t ₃ ) ends at which the oxygen reservoir is filled when it falls below a Threshold in an output signal of the post-cat lambda probe (6), characterized in that the integral obtained in this way is corrected, with a time offset (t ₁ - to) between the time of the change and a change in the sign of the slope of the output signal of the post-cat probe for the purpose of correcting (6) is measured and taken into account when correcting. Method for determining the oxygen storage capacity of an oxygen storage device (4) assigned to a catalytic converter (3) in an arrangement in which, in the outflow direction of the exhaust gas of an internal combustion engine (1), a pre-catalytic converter (5) in front of the catalytic converter (3) and behind at least a section of the catalytic converter ( 3) a post-catalytic converter (6) measures the air-fuel ratio, in which case the oxygen storage (4) is initially charged with lean exhaust gas under control using the pre-catalytic converter (5) in order to fill it with oxygen as much as possible , and then there is a change to a loading of the oxygen reservoir with rich exhaust gas in order to empty it of oxygen again, the amount of oxygen discharged per time interval being integrated over a time interval which begins at the point in time (t ₀ ) of the change and with a Point in time (t ₃ ) ends at which the oxygen reservoir is emptied when a s threshold value in an output signal of the post-catlambda probe (6), characterized in that the integral obtained in this way is corrected, with a time offset (t ₁ - t ₀ ) between the point in time of the change and a change in the sign of the slope of the output signal the post-cat probe (6) is measured and taken into account when correcting. Procedure according to Claim 1 or 2 , characterized in that intermediate values of the integral are held and the time offset to achieve a final value (t ₂ ) is subtracted from the time (t ₃ ) of the interval and the intermediate value assigned to the final value (t ₂ ) is used as a correction value. Procedure according to Claim 1 or 2 , characterized in that the proportion of calculated oxygen storage capacity is calculated which is included in the integral over a period of the length of the time offset before the end of the time interval, and that this proportion is subtracted from the integral to achieve a correction value. Procedure according to Claim 4 , characterized in that a change in the exhaust gas mass occurring during the time interval is taken into account when estimating the proportion. Procedure according to Claim 4 or 5 , characterized in that a change in the air-fuel ratio occurring during the time interval is taken into account when estimating the proportion. Procedure according to Claim 5 or 6 , characterized in that a filter constant for a digital low-pass filter is derived from the time offset and the oxygen input per time is filtered with the low-pass filter, a result of the filtering, in particular at the time of the end of the time interval, and the time offset being used when calculating the correction value will.

Images