AT516469B1 - METHOD FOR TARGETED GENERATION OF NH3 DURING THE REGENERATION PROCESS OF NOX MEMORY CATALYST - Google Patents

Abstract

The invention relates to a method for the specific generation of NH3 during the regeneration process of a in an exhaust system of an internal combustion engine (1) arranged NOX storage catalyst (LNT), wherein the regeneration is triggered by setting a reducing exhaust gas atmosphere before the NOX storage catalyst (LNT) and the time profile of at least during the regeneration phase upstream and downstream of the NOx storage catalytic converter (LNT) determined lambda values (λ1, λ2) in the exhaust gas upstream and downstream of the NOx storage catalyst (LNT) is determined. The following steps are carried out: a) Searching for a crossover point (K) at which the determined lambda values (λ1, λ2) have the same value upstream and downstream of the NOx storage catalytic converter (LNT); b) determining a differential area (D) spanned from the point of intersection (K) up to a reference time (TB) by the temporal profiles of the lambda values (λ1, λ2) upstream and downstream of the NOX storage catalytic converter (LNT); c) determining an amount of NH3 in the exhaust gas downstream of the NOX storage catalyst (LNT) based on the differential area (D); d) termination of the regeneration process, when the determined amount of NH3 in the exhaust downstream of the NOx storage catalyst (LNT) exceeds a defined limit.

Description

Description: The invention relates to a method for the targeted generation of NH ₃ during the regeneration process of an ΝΟχ storage catalytic converter arranged in an exhaust system of an internal combustion engine, the regeneration being triggered by setting a reducing exhaust gas atmosphere in front of the NO _x storage catalytic converter and the time course is determined by at least during the regeneration phase upstream and downstream of the ΝΟχ storage catalytic converter, the lambda values in the exhaust gas determined upstream and downstream of the ΝΟχ storage catalytic converter.

[0002] There are various possibilities for fulfilling current and future emission limit values when operating internal combustion engines mentioned at the beginning. A NO _x storage catalytic converter, which is also known as LNT (“lean NO _x trap), can be combined with an SCR catalytic converter, for the operation of which NH _{3 is} necessary, which can be formed in the LNT by prolonged rich operation. This internal NH ₃ generation during the extended LNT regeneration in combination with a downstream SCR is called a passive SCR. Alternatively or additionally, an active NH ₃ metering can be provided from the outside.

Usually, the end of the regeneration of a ΝΟχ storage catalytic converter, which is also known as LNT (Lean NO _X Trap), is recognized on the basis of a characteristic drop in the lambda value of a lambda probe arranged after the ΝΟχ storage catalytic converter, see WO 00/19075 A1. Alternatively, the point of intersection (lambda crossing) of the time profiles of the lambda values upstream and downstream of the ΝΟχ storage catalytic converter is used to identify the end of the regeneration (DE 10 2010 001 202 A1). Furthermore, it is known from DE 10 355 037 B4 to use a characteristic increase and subsequent decrease in a NO _x value of a NO _x sensor which is arranged downstream of the NO _x storage catalytic converter.

[0004] EP 2 149 684 A1 describes a method for controlling NO _x purification in an exhaust system of an internal combustion engine, with a reducing exhaust gas atmosphere being set upstream to regenerate the NO _x storage catalytic converter. In addition, the measured values of an upstream lambda sensor and other engine parameters are used to determine the NH ₃ generation from previously determined maps.

[0005] US 2013/0 019 589 A1 discloses an exhaust gas aftertreatment system for self-igniting internal combustion engines, which consists of a three-way catalytic converter element, a und storage catalytic converter and an oxidation catalytic converter element. It is provided that the NH3 saturation of the storage catalytic converter is determined using a model based on a number of operating parameters and measured values.

Known methods, however, have the disadvantage that they are relatively inaccurate and / or that additional and complex measuring sensors (e.g. at least one ΝΟχ-probe, which currently costs approximately three times the value of simple lambda probes) are required for the measurement.

It is the object of the invention to avoid these disadvantages and to propose a method which manages without additional expensive sensors and enables high-quality statements.

[0008] According to the invention, this is achieved by performing the following steps:

A) Searching for an intersection at which the determined lambda values upstream and downstream of the ΝΟχ storage catalytic converter have the same value, [0010] b) determining an from the intersection time to a reference time by the time profiles of the lambda values upstream and downstream of the ΝΟχ storage catalyst spanned differential area;

/ 7

AT516 469 B1 2018-06-15 Austrian Patent Office [0011] c) determining an amount of NH ₃ in the exhaust gas downstream of the NO _x storage catalytic converter based on the difference area;

D) Terminating the regeneration process when the determined amount of NH ₃ in the exhaust gas downstream of the ΝΟχ storage catalytic converter exceeds a defined limit.

The inventive method allows a conscious NH ₃ generation by defined extension of the regeneration process. The NH ₃ can then be stored in the exhaust gas aftertreatment system, for example a downstream SCR catalytic converter. The determination of the lambda values can be implemented with any probes that can measure lambda values - for example dedicated lambda probes or probes that measure O ₂ values, but also probes that determine the lambda value as an intermediate value. This means that no expensive NO _x probes with the corresponding NH ₃ cross sensitivity are required for the NH ₃ determination.

At the time of crossing in a), it is the time from or after which the lambda value downstream of the NO _x storage catalytic converter is smaller than the upstream lambda value. Before this crossing point, the downstream lambda value is greater than upstream.

The reference time in b) is the time of the end of regeneration, where the downstream lambda value rises again, in particular via the value at the time of crossing or above. This reference time is also characterized by the change in the upstream lambda value back to the normal operating value outside of the regeneration mode.

The limit value in d) is defined in particular by the amount of NH ₃ that is required in the downstream exhaust system, in particular to be stored in a downstream SCR catalytic converter.

[0017] In a first variant of the invention, in c), the NH ₃ amount determined based on a ₃ amount empirically determined relationship between the difference of surface and NH. The relationship is established in a known manner using an empirical formula or a corresponding map. The following parameters can be taken into account: temperature of the exhaust gas and the catalyst (or LNT); Exhaust gas mass flow; Engine operating points such as engine speed and load. The respective differences of the upstream and downstream lambda values are weighted with empirically determined factors and the weighted differences are integrated into the difference area. At any point in time, a lambda difference value is used to draw conclusions about an amount of NH ₃ , which is integrated into a total amount of NH _3, using the empirically determined relationship and the exhaust gas mass flow. In step d) the regeneration is stopped when a certain amount of NH _{3 has been} reached.

A higher-level SCR coordination thus ensures that sufficient NH _{3 is present} in the exhaust system or is stored in an SCR catalytic converter in order to ensure a desired NO _x conversion.

According to a second variant, a particularly simple determination of the NH ₃ amount in the exhaust gas results in c) if an H ₂ concentration in the exhaust gas downstream of the ΝΟχ storage catalytic converter is determined on the basis of the difference area and the NH ₃ amount in Exhaust gas downstream of the NO _x storage catalytic converter is determined on the basis of the H ₂ concentration determined. After the time of crossing, stored O _{2 is} used up and H ₂ formation occurs, which can be detected accordingly. This means that the lambda values upstream and downstream should be the same when all O _{2 has been} used up. The H ₂ formation takes place, for example, by splitting water molecules into H ₂ and ¹ / s O ₂ in the SCR catalytic converter. Because of the greater cross sensitivity of the sensor, with which lambda values are determined - in particular, for example, a lambda probe - for H ₂ than for O ₂ , the ratio of the concentrations of H ₂ to O ₂ in the probe is greater than in the exhaust gas in the catalytic converter and therefore the lambda value is smaller ,

2.7

AT516 469 B1 2018-06-15 Austrian Patent Office [0020] The determined NH ₃ quantity can be fed as an input variable to an NH ₃ loading model for an SCR catalyst (SCR = Selective Catalytic Reduction) through which the exhaust gas flows downstream of the NO _x storage catalyst ,

According to the invention, the NH ₃ is balanced based on the area of the lambda values spanned for and after the regeneration end before and after the NO _x storage catalytic converter and used in an NH ₃ loading model for a subsequent SCR catalytic converter.

The invention is based on the property of the empty ΝΟχ storage catalytic converter, with continued substoichiometric engine operation (rich operation, lambda value <1) to produce ammonia (NH ₃ ). On a ΝΟχ storage catalytic converter, H ₂ is primarily produced from CO primarily by the so-called water gas shift reaction, but also by steam reforming from HC. This H ₂ is used to reduce desorbed from the memory locations of the NO _x storage catalytic NO _x. As soon as no more NO _x is stored, or in some cases even before, H ₂ and NO NH _{3 are} formed.

The invention is based on the fact that after the regeneration of the ΝΟχ storage catalytic converter has ended - which is determined by means of the so-called lambda crossing - the lambda probe downstream of the ΝΟχ storage catalytic converter is used as an H ₂ sensor, in that the lambda values are based on the spanned differential area or O ₂ signals of the lambda sensors upstream and downstream of the ΝΟχ storage catalytic converter, the H ₂ concentration is calculated.

The invention makes use of the observation that the H ₂ concentration is proportional to the differential area D spanned by the time profiles of the lambda values upstream and downstream of the ΝΟχ storage catalytic converter:

D = K1 H ₂ .

The amount of NH ₃ formed is again proportional to the amount of H ₂ present:

H ₂ = K ₂ NH ₃ .

[0026] K1 and K2 are each proportionality factors. In addition, it is taken into account that the O ₂ signals of the lambda probes are more sensitive to H ₂ than to CO and HC and O ₂ .

The invention is explained in more detail with reference to a non-limiting embodiment in the figures.

1 [0030] FIG. 2 shows a schematic arrangement of an internal combustion engine for carrying out the method according to the invention, the time profiles of the lambda value, the H ₂ concentration and the NH ₃ amount in the exhaust gas downstream of the ΝΟχ- storage catalyst.

Fig. 1 shows schematically an internal combustion engine 1 with an exhaust system 2, in which a ΝΟχ storage catalyst LNT (LNT = Lean NO _X Trap), a particle filter cDPF and an SCR catalyst SCR are arranged to comply with the legal emission regulations. In the exemplary embodiment, the ΝΟχ storage catalytic converter and the particle filter cDPF are designed as a structural unit. Reference number 3 denotes the inlet line and reference number 4 denotes an exhaust gas recirculation system that branches off from the exhaust system 2 between the particle filter cDPF and the SCR catalytic converter and opens into the inlet system 3. However, the invention can be used not only in the low-pressure exhaust gas recirculation shown (removal of the exhaust gas upstream of an exhaust gas turbocharger), but also in high pressure exhaust gas recirculation (removal of the exhaust gas upstream of an exhaust gas turbocharger).

A first lambda sensor L1 is arranged upstream of the ΝΟχ storage catalytic converter and a second lambda sensor L2 is arranged downstream of the ΝΟχ storage catalytic converter LNT.

3.7

AT516 469 B1 2018-06-15 Austrian patent office

Basically, any sensors can be used with which a lambda value can be determined - lambda probes according to the exemplary embodiment described are one of several possibilities. The lambda sensors (lambda sensors) L1, L2 measure the residual oxygen content O ₂ in the exhaust gas, which results after combustion of all previously unburned fuel components (e.g. HC, CO) in order to be able to adjust the ratio (lambda value) λ of combustion air to fuel. If there is insufficient oxygen for this oxidation, it is pumped out of the exhaust gas in the opposite direction by an oxygen pump integrated in the lambda probe. This negative oxygen pump current is calculated as a negative oxygen concentration and subsequently as a lambda less than 1. The lambda value serves as a control sensor for a lambda control loop (not shown further).

The NO _x storage catalytic converter LNT serves for the temporary storage of nitrogen oxides (NO _x ) contained in the exhaust gas. To temporarily store the nitrogen oxides, the NO _x storage catalyst LNT has a noble metal catalyst such as platinum and a NO _x storage component, for example barium, on suitable supports. In the lean, that is, oxygen-rich, atmosphere, the nitrogen oxides are oxidized under the action of the noble metal catalyst, absorbed with the formation of nitrates such as barium nitrate in the catalyst, and thus removed from the exhaust gas stream. Through regular short-term enrichment of the exhaust gas mixture, i.e. lowering the lambda value λ-, these reactions take place in the opposite direction, as a result of which the NO _x molecules are released into the exhaust gas stream again and due to the reducing components present in the rich atmosphere, such as CmHn, incompletely burned hydrocarbons - and / or CO can be further reduced.

If the absorption capacity of the NO _x storage catalytic converter LNT is exhausted, the engine electronics briefly set a rich, reducing exhaust gas mixture for a few seconds (approximately two to ten seconds). In this short, rich cycle, the nitrogen oxides NO _x temporarily stored in the catalytic converter are reduced to nitrogen N ₂ and the NO _x storage catalytic converter is thus prepared for the next storage cycle. In order to keep engine operation as short as possible in the rich cycle, it is essential in classic LNT operation that the end of reduction is recognized as precisely as possible.

In the downstream SCR catalytic converter, NO _x molecules are reduced by NH ₃ to N ₂ in the case of passive SCR for further nitrogen oxide reduction. The amount of NH _{3 to} be provided is determined using an SCR catalyst model.

Fig. 2 shows the curves of the lambda value Aj upstream of the NO _x storage catalytic converter LNT, the lambda value λ ₂ downstream of the NO _x storage catalytic converter LNT, the course of the proportions of nitrogen oxide NO _x downstream of the NO _x storage catalytic converter LNT, the course the proportions of hydrogen H ₂ downstream of the NO _x storage catalytic converter, the course of the proportions of ammonia NH ₃ downstream of the NO _x storage catalytic converter LNT, the course of the proportions of carbon monoxide CO downstream of the NO _x storage catalytic converter LNT and the course of the proportions of unburned Hydrocarbons HC downstream of the NO _x storage catalytic converter LNT, in particular during a regeneration phase REG of the NO _x storage catalytic converter LNT plotted over time t.

During the regeneration phase REG, the internal combustion engine 1 is operated substoichiometrically, as can be seen from the line M indicating the operating mode. 1 stands for rich operation and 2 for lean operation. At the point marked with reference number R1, the change from lean to rich engine operation takes place.

[0038] The regeneration phase REG is divided into two time segments. The first time period A is characterized in that the lambda value Aj upstream of the NO _x storage catalytic converter LNT is smaller than the lambda value λ ₂ downstream of the NO _x storage catalytic converter LNT. In the second period, however, the lambda value Aj upstream of the NO _X storage catalytic converter LNT is greater than the lambda value λ ₂ downstream of the NO _X storage catalytic converter LNT. At the point of intersection K the courses of the two lambda values A _1; λ ₂ indicate the lambda values λ ₁ determined by means of the lambda sensors L1, L2 _; λ ₂ upstream and downstream of the NO _x catalyst to the same value.

4.7

AT516 469 B1 2018-06-15 Austrian Patent Office Surprisingly, it was found that there is a relationship between the difference area D between the courses of the lambda values λ _1; λ ₂ , starting with the crossing time K and ending with the reference time T _B , and the NH ₃ generated during the regeneration process. In a first variant of the invention, therefore, an empirically determined connection is used, which is established by an empirical formula or a corresponding map. The following parameters can be taken into account: temperature of exhaust gas and / or catalytic converter (eg LNT); Exhaust gas mass flow; Engine operating parameters such as engine speed and load.

The lambda differences are weighted with empirically determined factors and the weighted differences are integrated into the difference area D. After a desired amount of NH _{3 has been reached} downstream, the regeneration is stopped.

[0041] A further variant of the invention is as follows: As soon as all cached nitrogen oxides NO _{x is} reduced to nitrogen N ₂ and the NO _x storage catalytic converter LNT is therefore empty, is produced in this hydrogen H _2, indicated by reference numeral R2. The proportion of hydrogen H ₂ is usually more than ten times the proportion of carbon monoxide CO, which is again more than ten times the proportion of unburned hydrocarbons HC (comparison R3 and R4). Upstream of the NO _x storage catalytic converter LNT, hardly any H _{2 can be found} (or is immediately oxidized with the residual oxygen or the O ₂ stored in the LNT). Studies have shown that the hydrogen concentration H ₂ generated in the second period B of the regeneration phase REG is proportional to the difference area D between the courses of the two lambda values λ _1; λ ₂ - starting at the time of crossing K and ending at the reference time T _B - is.

Since the differences in the lambda values are thus primarily determined by the hydrogen H ₂ , the lambda sensor L1, L2 can be used as a hydrogen sensor. Furthermore, different sensitivities of the lambda sensor to carbon monoxide CO and hydrocarbons HC are known.

The concentration of hydrogen H ₂ can thus be set as being proportional to the difference area D:

D = K1 H ₂ , where K1 represents a first proportionality factor.

The production of the hydrogen is primarily due to the so-called water gas shift reaction H ₂ from CO H ₂ O + CO <-> CO ₂ + H ₂ but also to steam reforming from HC:

3H ₂ O + C ₃ H ₆ <-> 3CO + 6H ₂ .

[0045] This H ₂ is used to reduce desorbed from the memory locations of the NO _x storage catalytic NO _x. As soon as no more NO _x is stored, or in some cases even before, H ₂ and NO NH _{3 are} formed:

8H ₂ + Ba (NO ₃ ) ₂ -> 5H ₂ O + BaO + 2NH ₃ .

The amount of NH3 formed is again proportional to the amount of H ₂ present:

H ₂ = K2 NH ₃ , where K2 represents a second proportionality factor.

The amount of NH ₃ determined in this way can be fed as an input variable to a known NH ₃ loading model for an SCR catalyst SCR through which the exhaust gas flows, downstream of the NO _x storage catalytic converter.

5.7

AT516 469 B1 2018-06-15 Austrian

Patent Office

Claims (6)

claims 1. A method for the targeted generation of NH ₃ during the regeneration process of a NO _x storage catalytic converter (LNT) arranged in an exhaust system of an internal combustion engine (1), the regeneration being triggered by setting a reducing exhaust gas atmosphere in front of the NO _x storage catalytic converter (LNT) and the time profile of lambda values (λ _1; λ ₂ ) determined in the exhaust gas upstream and downstream of the NO _x storage catalytic converter (LNT) at least during the regeneration phase upstream and downstream of the NO _x storage catalytic converter (LNT), characterized in that the following steps are carried out: a) searching for an intersection point (K) at which the determined lambda values (λ _1; λ ₂ ) upstream and downstream of the NO _x storage catalytic converter (LNT) have the same value; b) determining a difference area (D) spanned from the time of intersection (K) to a reference time (T _B ) by the time profiles of the lambda values (λ _1; λ ₂ ) upstream and downstream of the NO _x storage catalytic converter (LNT); c) determining an amount of NH ₃ in the exhaust gas downstream of the NO _x storage catalytic converter (LNT) on the basis of the difference area (D); d) termination of the regeneration process when the determined amount of NH ₃ in the exhaust gas downstream of the NO _x storage catalytic converter (LNT) exceeds a defined limit value. 2. The method according to claim 1, characterized in that in step c) the amount of NH _{3 is} determined on the basis of an empirically determined relationship between the difference area (D) and the amount of NH ₃ . 3. The method according to claim 1, characterized in that in step c) an H ₂ concentration in the exhaust gas downstream of the NO _x storage catalytic converter (LNT) is determined on the basis of the difference area (D) and the amount of NH ₃ in the exhaust gas downstream of the NO _X storage catalytic converter (LNT) is determined on the basis of the determined H ₂ concentration. 4. The method according to claim 3, characterized in that the H ₂ concentration in the exhaust gas downstream of the NO _x storage catalytic converter (LNT) is assumed to be proportional to the difference area (D): D = K1 H ₂ , where K1 is a first proportionality factor. 5. The method according to claim 3 or 4, characterized in that the amount of NH ₃ in the exhaust gas downstream of the NO _x storage catalyst (LNT) is assumed proportional to the hydrogen concentration H ₂ : H ₂ = K2 NH ₃ where K2 is a second proportionality factor. 6. The method according to any one of claims 1 to 5, characterized in that the NH determined ₃ -Beladungsmodell for a downstream of the NO _x storage catalytic converter (LNT) perfused ₃ quantity as input variable to a NH from the exhaust SCR catalyst (SCR) is fed ,