WO2016146907A1

WO2016146907A1 - Method for purging or charging oxygen of a catalyst installed in the flow of an engine

Info

Publication number: WO2016146907A1
Application number: PCT/FR2016/050371
Authority: WO
Inventors: Damien Llory; Damien Lefebvre; Nils Matthess; Frederic Dambricourt
Original assignee: Peugeot Citroen Automobiles Sa
Priority date: 2015-03-17
Filing date: 2016-02-18
Publication date: 2016-09-22
Also published as: FR3033832A1; FR3033832B1

Abstract

The invention concerns a method for purging or charging oxygen of a catalyst arranged in the flow of an exhaust gas purification system of an engine comprising the catalyst, a gas probe capable of supplying a signal making it possible to determine the oxygen content of the exhaust gases or the richness of the exhaust gases, downstream from the catalyst, the method comprising: a step of purging (B') - or respectively charging - oxygen in the catalyst, characterised in that the time of completion of step (B') is validated by an expected response of the signal (20) from the probe downstream from the catalyst caused by an upward - or respectively downward - pulse (30) of richness during the first step (B').

Description

PROCESS FOR PURGING OR OXYGEN LOADING OF A CATALYST INSTALLED IN THE FLOW OF AN ENGINE

The present invention relates to a method for diagnosing a catalyst installed in the flow of an internal combustion engine, and more specifically relates to the measurement of the oxygen storage capacity of such a catalyst. This process is part of the field of exhaust gas monitoring and regulation to optimize their composition for purification of these gases.

In this field, the optimal composition of the exhaust gas of a motor vehicle is generally obtained by continuous monitoring of oxygen and reducing species such as hydrocarbons and carbon monoxide emitted. In fact, the harmful exhaust gases to be eliminated are the remaining hydrocarbons, carbon monoxide and nitrogen oxides. However, the oxidation reactions of the hydrocarbons and carbon monoxide require the presence of oxidizing species, in particular oxygen, whereas the reduction reaction of the nitrogen oxides requires the presence of reducing species such as hydrocarbons or carbon monoxide.

The catalytic purification of harmful exhaust gases is generally carried out with a "three-way" catalyst (abbreviated TWC, initials of "Three-Way-Catalyst" in English terminology), which transforms both the three pollutants, namely hydrocarbons and carbon monoxide by oxidation, as well as nitrogen oxides by reduction. The yield of these reactions is maximum when the catalyst has as many oxidizing species as reducing species. We then speak of stoichiometric conditions or of the richness of the injected air - fuel mixture equal to 1. In practice, there is a wealth window of this TWC catalyst around the value 1 where the reaction efficiency is optimal.

[0004] Stoichiometric conditions of richness (hereinafter also referred to as "R") equal to 1 also correspond to a value of 1 for a lambda coefficient "Λ", the inverse of "R". This is why the probes generally used for this type of measurement are called "lambda probes". The magnitude "wealth" and the size "lambda" are both representative of the relative proportions of oxygen and reducing species such as hydrocarbons in the exhaust gas, their values moving in the opposite direction. Specifically, the "R" richness formula is the ratio between fuel injected and stoichiometric mixture fuel proportions (eg, 14.5 g of air is required to burn 1 g of gasoline) compared to the corresponding air mass, namely:

Mci / M _Ai

R =

Mes / M _As

With: M _C i = fuel mass actually injected, M _A i = air mass actually injected, M _Cs = fuel mass in stoichiometric mixture, and M _As = mass of air in stoichiometric mixture.

The richness around the stoichiometry over time is regulated with a frequency that can vary from about 1 to 4 Hz, and with an amplitude of ± 3% of wealth approximately. This regulation is provided by a lambda probe positioned upstream of the catalyst, which sends a signal to the engine control to indicate whether it should increase or reduce the injection time to increase or decrease the wealth. Another lambda probe, placed downstream of the catalyst, then verifies its effectiveness especially for on-board diagnostics (for example, the EOBD, according to the "European On Board Diagnostics" standard in English terminology).

Having defined the quantities involved in the exhaust gas, it should be noted that the evolution of the efficiency of a catalyst - that is to say, its ability to transform harmful gases - is appreciated by its oxygen storage capacity, called OSC (initials of "Oxygen Storage Capacity" according to English terminology). It is therefore recommended to accurately measure the OSC of the catalyst and adjust the value of "R".

[0007] Different methods are used to measure the OSC of a catalyst. A basic method of this measurement, presented with reference to the diagram of Figure 1, involves an upstream probe and a downstream probe disposed at the ends of the catalyst tested. The diagram of Figure 1 presents the simultaneous evolution of three quantities during the time "t" concerning these probes and the catalyst: the variation curve 1 of the voltage supplied by the probe placed upstream of the catalyst, the curve 2 of variation of the voltage supplied by the probe placed downstream of the catalyst, and the variations 3 of the richness "R" of the mixture injected into the engine.

The process consists in purging the catalyst of its oxygen (thanks to a purge phase) and then loading it again with oxygen (thanks to a storage phase) by two successive injections of respectively richer and poorer content. fuel only in the stoichiometric mixture. The calculation of the OSC is conducted during this storage phase. Oxygen storage capacity, OSC is determined by the integral over the storage duration of the oxygen flow rate:

With m ₀₂ the oxygen flow rate or the integral over the storage duration of a function dependent on the richness R and the exhaust gas flow rate:

OSC = {(1 - R) · K · m _{exhaust gas} dt

[0012] With R wealth, m _{gazéchappement} the flow of exhaust gas, and

K a correction coefficient.

In the example, the vertical axis represents the magnitude "V" in Volt voltages of the probes and the dimensionless value of the wealth "R". The intersection of this axis "V" and the axis of the time scale "t" corresponds to both a richness "R" equal to 1 (stoichiometric ratio between oxygen and reducing species such as hydrocarbons ) and at a base value of the voltages here equal to 0.6 V.

In the basic regime, illustrated by a first period of time "A" of richness "R" of base equal to 1 at the catalyst inlet, the signal 1 of the upstream probe oscillates around the stoichiometry and the signal 2 of the downstream probe is smoothed. In fact, the catalyst components, mixed oxides of cerium and zirconium, buffer the wealth oscillations detected upstream thanks to very fast oxidation and reduction reactions. For example, for cerium, the fast oxidation-reduction formulas are of the type:

This ₂ 0 ₃ + ½ 0 ₂ -> 2 Ce0 ₂

and

2 Ce0 ₂ -> Ce ₂ 0 ₃ + 1/2 0 ₂

The second period of time "B" purge phase begins with a gas injection of a level of wealth, which will be referred to as purge wealth, higher than the basic wealth. The gas injection is for example richer in fuel by about 5% (and therefore depleted of oxygen), which results in a rising niche of the curve 3 of richness passing from 1 to 1.05. This phase B allows the catalyst to be purged of oxygen that could possibly contain. This purge causes the release of the oxygen contained in the mixed oxides of the catalyst generating their reduction. The oxygen thus released is then used to oxidize the reducing species such as hydrocarbons. The signal 1 of the upstream sensor goes substantially to its maximum in the upstream slot. And the signal 2 of the downstream probe weakly increases at the end of phase "B".

At the end of this purge phase "B", where it is assumed that the catalyst has completely purged its oxygen, a charge phase "C" (also called storage phase), is triggered by the injection of an excess of oxygen. The measurement of the OSC of the catalyst is carried out during this charging phase "C".

At the beginning of this "C" phase, the excess oxygen supply corresponds to an instantaneous drop in the "R" richness increasing its value from 1.05 to 0.95 (curve 3). The richness curve 3 then forms a falling slot during the duration of the charging phase "C".

The signal 1 of the upstream probe also drops sharply, in the downlink slot, but with a delay time dT, while the signal 2 of the downstream probe remains substantially at the high level it had in the purge phase. "B" because the catalyst substantially recovers all the excess oxygen. This situation lasts during a charging period "T" required for the catalyst to charge oxygen. When the catalyst becomes saturated with oxygen at the end of period "T", the excess oxygen exits the catalyst, which suddenly drop the level of the signal 2 of the downstream probe. Once the charging phase "C" is complete, the basic regime of the period "A" resumes: the injection is adjusted to the stoichiometry (the level of richness "R" returns equal to 1), the signal 1 of the The upstream probe resumes its oscillations around the stoichiometry, and the signal 2 of the downstream probe progressively rises towards the value of the basic regime (phase "A"). The end of the charging period "T" is generally defined at the moment of the regain of the basic speed or during the fall of the signal 2.

The measurement of the OSC of the catalyst is then calculated by the integral over the duration T of the charging period according to one of the preceding relationships.

A main disadvantage of this method is that it does not validate with certainty the end of the oxygen purge of the catalyst in the "B" phase, which generally falsifies the calculation of the OSC. Now it is essential to monitor directly and precisely the oxygen storage capacity of the catalyst, because it is a key marker of aging.

It was therefore tried to improve this measure of the OSC, as evidenced by the solutions developed in the documents below. For example, the patent document FR2798700 teaches the possibility of reaching an estimate of the oxygen storage capacity, OSC, of the catalyst from the delay time that the oxygen concentration at the catalyst outlet makes to rise above the oxygen concentration. a certain threshold after a decline in wealth. This estimate taking into account the air pressure at the engine inlet, the engine speed and a correspondence table between said pressure and said engine speed. The estimated OSC value must then be compared to a reference curve before concluding the performance of the catalytic converters observed. This CSO estimate is therefore complex and does not provide a reliable and accurate OSC measurement value.

Moreover, the patent document FR2858019 presents a measurement method in a system comprising two catalysts in parallel, with a simultaneous action of the measurement of the OSC and other actions or measurements, concerning for example the regeneration of the oxides of nitrogen NOx, the plausibility control of the information given by the probes, or the study of the dynamic behavior of the probes, such as the study of the gradient of the behavior over time of a lambda probe. This solution also provides a complex estimation method of OSC.

In addition, the patent document FR2910052 proposes to judge the properties of the catalyst using the composition of the downstream emission of the catalyst to associate directly with the quality of the catalyst. In the gases observed by the downstream probe, the gases H ₂ (dihydrogen) and CO (carbon monoxide) are present in a proportion which is a function of the quality of the catalyst, and this measurement of the proportion of these gases by a probe makes it possible, indirectly, to determine the quality of this catalyst. This indirect determination to calculate the OSC remains unreliable.

The main purpose of the invention is therefore to provide a reliable determination of the OSC of a catalyst - its oxygen storage capacity - by a simple, fast, accurate and reliable method, by only involving the parameters probes placed upstream and downstream of the catalyst. For this purpose, the invention proposes to exploit the specific response of the downstream probe to variations in the richness of the exhaust gases.

More specifically, the subject of the present invention is a process for purging or oxygen charging a catalyst arranged in the flow of an exhaust gas purification plant of an engine comprising the catalyst, a gas probe capable of providing a signal for determining the oxygen content of the exhaust gas or the richness of the exhaust gas downstream of the catalyst, the process comprising:

a first step of purging - respectively loading - of oxygen in the catalyst,

characterized in that the end time of this first step is validated by an expected response of the signal of the downstream catalyst probe caused by an upward or downward pulse of richness during the first step.

Various additional features may be provided, alone or in combination: Several upward or downwardly increasing pulses of richness are successively injected until a pulse provokes the expected response of the signal from the downstream probe.

The upward pulses - respectively downward - successive richness succeed each other with a variable frequency and / or a variable amplitude.

The first purge step - respectively charge - of oxygen in the catalyst is followed by a second storage step - respectively purge oxygen in the catalyst.

Determining the partial oxygen storage capacity and / or total during the second step.

The catalyst is a "three-way" catalyst.

The gas probe downstream of the catalyst is chosen between a lambda probe or a nitrogen oxide probe.

When the purification plant comprises several catalysts in series, the application of the step of determining the oxygen storage capacity is chosen between an individualized determination dedicated to each of the catalysts, a group determination dedicated to a catalyst group and a global determination dedicated to all the catalysts.

The invention also relates to a control unit comprising the acquisition means, processing by software instructions stored in a memory and the control means required for the implementation of the method according to any one of the previously described variants.

The invention also relates to an engine comprising a plant for purifying the flow of the exhaust gases produced by said engine in which a catalyst is arranged, characterized in that it comprises such a control unit for the implementation of the method of the invention.

PRESENTATION OF FIGURES Other data, features and advantages of the present invention will appear on reading the following nonlimited description, with reference to the appended figures which represent, respectively:

FIG. 1, a measurement method for determining the OSC of a catalyst according to the state of the art (commented on above);

FIG. 2, an exemplary implementation of implementation of the method of the invention;

FIG. 3a, an example of a measurement diagram for the determination of the OSC of a catalyst according to the invention, in the case where the first step is a purge of the catalyst and the catalyst is completely purged of oxygen at moment of injection of an impulse increase of wealth;

FIG. 3b, an example of a measurement diagram for the determination of the OSC of a catalyst according to the invention, in the case where the first step is a purge of the catalyst and the catalyst is not completely purged of oxygen at the time of the first injection of an impulse increase of richness, and

FIG. 3c, an example of a measurement diagram for the determination of the OSC of a catalyst according to the invention, in the case where the first step is a storage of oxygen in the catalyst and in which the catalyst is totally saturated with oxygen at the moment of the injection of an impulse drop of wealth.

DETAILED DESCRIPTION

The axes of the diagrams of Figures 3a, 3b and 3c show the same magnitudes as those of Figure 1 and are designated by the same references. Any reference to an element refers to the passage of the description that details this element. In particular, the references to the mounting elements in the passages corresponding to the description of FIGS. 3a to 3c refer to the description of FIG. 2 of assembly.

Figure 2 shows an example of mounting that can be used to implement the invention. The catalyst 8 under OSC diagnostics is installed in the flow of an installation (not shown) for cleaning the exhaust gases of an internal combustion engine (not shown). The incoming flow 6E of the exhaust gas baths the end of an upstream lambda probe 7A. And the outflow 6S bathes the end of a downstream lambda probe 7B, placed downstream of the catalyst 8.

The upstream sensors 7A and downstream 7B are connected to a control unit adapted to control and regulate the motor 5 (hereinafter, unit 5). This control unit 5 comprises the acquisition means, software instructions processing stored in a memory and the control means required to implement the method of the invention described in this memory.

This unit 5 controls the richness, R, of the air-fuel mixture to be injected at any moment into the engine. Under the operating conditions of the basic speed (see phase A of FIG. 1), the proportion of fuel and air in the mixture is calculated by unit 5 so that R can be distinguished by the designation of base richness is substantially equal to 1. Each lambda probe 7a, 7b provides reference signal values at the unit 5, values which directly translate the oxygen content of the exhaust gas streams 6E and 6S and, consequently, the richness in these gases.

The catalyst 8 is designed to eliminate the polluting gases - the outgoing hydrocarbons, nitrogen oxides and carbon monoxide - and its effectiveness over time is measured by its OSC, that is to say its ability to fix oxygen, as has already been reported (with reference to Figure 1). When an OSC measurement of the catalyst is started, the unit 5 adjusts the injectors (not shown) of the mixture in the engine so as to cause predetermined wealth injection pulses. The injection pulse may for example be in the form of a slot of wealth.

The variations in the voltage of the probe 7b which result therefrom are then a function of the oxygen retention capacity of the catalyst 8, which will make it possible to determine the OSC of the catalyst to be diagnosed in the manner that will be exhibited. with reference to the diagrams of Figures 3a, 3b and 4.

FIG. 3a presents a first measurement diagram for the determination of the OSC of the catalyst 8, with a gas injection pulse 30 at a richness, called pulse richness, higher than the purge richness during the step B 'oxygen purge catalyst. The "A" phase of diet base and purge steps "B" by increasing oxygen richness and oxygen storage C correspond respectively to the "A" phase, as well as to the "B" and "C" purge phases described with reference in Figure 1.

In this Figure 3a are more precisely shown the evolution of the richness of the gas on the curve 3, as well as the evolution of the voltage 2 'of the lambda probe downstream 7B. A crucial point in the determination of the OSC is to be able to accurately measure the end of purge time, in other words the moment when the exhaust gases no longer contain any oxygen downstream of the catalyst. 8. Indeed, it is from this moment that the determination of the OSC must be started in order to obtain a reliable, fair and precise value.

[0047] According to the invention, an end-of-purge "M" marking of the catalyst 8 is carried out by a punctual increase in the richness of the mixture, carried out by a more fuel-rich injection pulse triggered by the unit 5. In the example, a 1% increase in wealth is injected. When the catalyst is completely purged of its oxygen, the pointwise increase in richness 30, which can be seen as a stimulus, then causes an expected response of the downstream probe signal 7B, here an increase of the voltage signal of the probe. downstream 7B.

This expected increase in the voltage of the downstream probe 7B validates the end of the purge of the catalyst 8 to the extent that the one-off decrease in oxygen, corresponding to the one-off increase in richness, can not then be compensated by a destocking of oxygen. Under these conditions, the duration T 'of the storage step C for determining the OSC of the catalyst 8 is initiated just after the detection of this notable variation and ends at the moment when the fall 21 of the curve of voltage 2 'is detected.

The diagram of FIG. 3b illustrates measurements of the assembly of FIG. 2, with a catalyst 8 not totally purged of oxygen when triggering the increased richness injection 31. This first marking test M 'is carried out during the purging step referenced B ", corresponding to the purge step B' described with reference to FIG. 3a, and which starts with the same enrichment of the mixture according to curve 3 ". In this case, the point increase in richness 31 is not followed by the expected increase indicative of a complete oxygen purge of the voltage 2 "of the downstream probe 7B, contrary to the increase 21. In these conditions, following the richness niche 31, the sending of the rich mixture 3 "at the so-called purge richness level continues during the purge phase B".

Then a new point increase 32 wealth is injected by the unit 5 in a pulse manner. This time, a significant increase 22 in the voltage curve 2 "of the downstream probe 7B is detected by the unit 5. This variation then validates the end of the purge step B" of the catalyst. The determination of the OSC is conducted by integrating the duration T "of the oxygen storage step C" as in the preceding example, that is to say between the moment of significant increase of voltage 22 of the voltage curve 2 "and the significant decrease time 23 of this same curve 2" at the end of storage step C ".

More generally, when a new richness pulse does not cause the expected response of the voltage curve 2 ", new injections are triggered injected by the unit 5 until the expected response of the curve is obtained. 2 "voltage. Advantageously, these new injections are successively provoked with a variable frequency and / or a variable amplitude, for example increased, so as to optimize the detection of the significant rise in the voltage curve 2 ".

The diagram of Figure 3c shows another embodiment of the OSC measurement of the catalyst according to the invention. Contrary to the order made in the previous examples, in which the first step is a purge step followed by a second oxygen storage step, the present example starts with a step B2 for storing oxygen in the catalyst followed by a purge step C2. The basic diet phase "A" is unchanged.

In this example, the catalyst 8 is completely full of oxygen in the oxygen storage step B2 at the moment when a mixture injection pulse corresponding to a one-off reduction of the 3 "'wealth relative to the so-called charge wealth is triggered by unit 5. The decline punctual wealth then causes an expected response, in this case a decrease 25, of the voltage curve 2 "of the downstream probe 7B.The injection of the pointwise drop of richness 35 forms an M2 marker of end of oxygen charge and the moment of detection of the significant decrease which validates this end of oxygen storage of the catalyst 8. Step C2 then corresponds to the purge of oxygen caused by the rise in richness 3 ".

The end of the purge step C2 coincides with the rise of the voltage 2 "'of the downstream probe 7 B. The duration T2 of the purge step C2 extends more precisely between the detection instant of the significant decrease 25 and the detection time of the ascent 26 of the voltage curve 2 "'. The integration of this value of the duration T2 in the triple integration technique mentioned above then makes it possible to determine the OSC of the catalyst to be diagnosed.

The invention is not limited to the embodiments described and shown. It is for example possible to also determine the partial OSC, for example by ceasing the calculation when the value of the integral reaches a determined oxygen threshold which is lower than the total OSC of the catalyst.

Alternatively, it is possible to use a downstream probe "on / off" (binary oscillation), or proportional. More generally, any type of sensor capable of providing a signal for monitoring the oxygen content of the exhaust gas or the richness of the exhaust gas in the mixture to be injected into the engine can be used as a probe. Thus, a lambda probe or a NOx could be suitable. The upstream sensor can also be replaced by an estimator.

In another variant, in which the purification plant comprises several catalysts 8 in series, the application of the partial and / or total OSC determination determining step is chosen between a dedicated individualized determination. to each of the catalysts 8, a group determination dedicated to a group of catalysts 8 and a global determination dedicated to all the catalysts 8.

Furthermore, the instant of end of duration of the second stage (storage or purge) corresponding to the determination of the OSC of the catalyst to be controlled, can be chosen between the moment of ascent or descent of the curve voltage of the downstream probe, as in the examples above, but also at the moment of control of the change of richness, in order to return to basic speed, or at the moment when the voltage of the downstream probe passes by the base value.

Claims

1. Process for purging or oxygen loading a catalyst (8) arranged in the flow of an exhaust gas purification plant of an engine comprising the catalyst (8), a gas probe capable of supplying a signal for determining the oxygen content of the exhaust gas or the richness of the exhaust gas downstream (7B) of the catalyst (8), the method comprising:

a first purge step (Β 'B ") - respectively charge (B2) - of oxygen in the catalyst (8),

characterized in that the end time of this first step (B '; B "; B2) is enabled by an expected response of the signal (20; 22; 25) of the downstream probe (7B) of the catalyst (8) caused by an upward (respectively downward) (30) boost of richness during the first step (B '; B "; B2).

2. The method according to claim 1, wherein a plurality of upward-downward-richness pulses are successively (31; 32) injected until a pulse causes the expected response (22) of the signal of the downstream probe (7B).

3. Method according to claim 2, wherein the upward-downwardly-successive pulses of richness (31; 32) succeed one another with a variable frequency and / or a variable amplitude.

4. Process according to any one of the preceding claims, wherein the catalyst (8) is a "three-way" catalyst.

The method of any one of the preceding claims, wherein the downstream gas probe (7B) of the catalyst (8) is selected from a lambda probe or a nitrogen oxide probe.

6. A method according to any one of the preceding claims, wherein the first purge step (B '; B ") - respectively charge (B2) - oxygen in the catalyst (8) is followed by a second step storage (C; C ") - respectively purge (C2) of oxygen in the catalyst (8).

7. The method of claim 6, wherein the partial and / or total oxygen storage capacity is determined during the second step.

8. The method of claim 7, wherein, the purification plant comprises several catalysts (8) in series, the application of the step of determining the oxygen storage capacity is chosen between an individualized determination dedicated to each of the catalysts (8), a group determination dedicated to a group of catalysts (8) and a global determination dedicated to all the catalysts (8).

9. Control unit (5) comprising the acquisition means, processing by software instructions stored in a memory and the control means required to implement the method according to any one of the preceding claims.

10. Motor comprising an installation for purifying the flow of the exhaust gases produced by said engine in which a catalyst is arranged, characterized in that it comprises a control unit according to claim 9 for implementing the method according to any of claims 1 to 8.