METHOD FOR MONITORING A NITROGEN OXIDE STORAGE CATALYST

Abstract

A method for monitoring a nitrogen oxide storage catalyst in an exhaust system of an internal combustion engine, in which a reduction of nitrogen oxides is carried out by means of a reducing agent is disclosed. During a regeneration of the nitrogen oxide storage catalyst, the following steps are carried out: A measurement is carried out, from which a slip rate of the reducing agent not absorbed in the nitrogen oxide storage catalyst is ascertained. In addition, at least one expected value for the slip rate of the reducing agent is ascertained from at least one model. Subsequently, a computation of a monitoring variable is carried out by means of the slip rate of the reducing agent ascertained from the measurement and the at least one expected value for the slip rate of the reducing agent. Finally, a diagnosis of the storage capacity of the nitrogen oxide storage catalyst is carried out on the basis of the monitoring variable.

Claims

1. A method for monitoring a nitrogen oxide storage catalyst (3) in an exhaust system (2) of an internal combustion engine (1), in which a reduction of nitrogen oxides is carried out by means of a reducing agent, and performed during a regeneration of the nitrogen oxide storage catalyst (3), the method comprising: carrying out a measurement (10), from which a slip rate (Q.sub.meas) of the reducing agent not absorbed in the nitrogen oxide storage catalyst (3) is ascertained; ascertaining (61, 62) at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent from at least one model (20, 30); determining, via a computer, (100) a monitoring variable (Q.sub.norm)based on the slip rate (Q.sub.meas) of the reducing agent ascertained from the measurement (10) and the at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent; and diagnosing (110), via the computer, the storage capacity of the nitrogen oxide storage catalyst (3) based on the monitoring variable (Q.sub.norm).

2. The method according to claim 1, wherein one of the expected values (Q.sub.WPA) for the slip rate of the reducing agent is ascertained from a model (20) for a WPA pattern, which represents an intact nitrogen oxide storage catalyst (3).

3. The method according to claim 1, wherein one of the expected values (E.sub.BPU) for the slip rate of the reducing agent is ascertained from a model (30) for a BPU pattern, which represents a defective nitrogen oxide storage catalyst (3).

4. The method according to claim 1, wherein the at least one expected value for the integral reducing agent mass (E.sub.WPA, E.sub.BPU) is ascertained by means of a multiplication (51, 52) of a slip rate (Q.sub.WPA, Q.sub.BPU) for the reducing agent ascertained from the at least one model (20, 30) with an available reducing agent flow (Q.sub.Ra) and subsequent integration (61, 62) over time.

5. The method according to claim 1, wherein, for each of the at least one expected values for the integral reducing agent mass (E.sub.WPA, E.sub.BPU), a mean value is ascertained for the slip rate of the reducing agent (Q.sub.WPA, Q.sub.BPU) by means of a division (71, 72) by an available reducing agent mass flow (Q.sub.Ra), and the monitoring variable (Q.sub.nom) is computed by means of the at least one mean value (Q.sub.WPA, Q.sub.BPU) and a mean slip rate (Q.sub.meas) of the reducing agent ascertained from the measurement.

6. The method according to claim 1, wherein at least one of the following parameters is used in at least one model (20, 30): a combustion air ratio (.sub.v) upstream of the nitrogen oxide storage catalyst (3); a setpoint value (.sub.s) of the combustion air ratio (.sub.v) upstream of the nitrogen oxide storage catalyst (3); a deviation (A.sub.vs) between the combustion air ratio (.sub.v) upstream of the nitrogen oxide storage catalyst and the setpoint value (.sub.s); the exhaust gas mass flow (Q.sub.A); the temperature (T.sub.cat) of the nitrogen oxide storage catalyst (3); the running consumption (m.sub.V) of the reducing agent since regeneration start as a measure of the regeneration progress; a nitrogen oxide load (m.sub.catS) of the nitrogen oxide storage catalyst (3); and a sulfur load (m.sub.catNOx) of the nitrogen oxide storage catalyst (3).

7. The method according to claim 6, wherein the parameters are incorporated into the model (20, 30) as characteristic maps (210) or characteristic curves (211-215).

8. The method according to claim 1, wherein the measurement (10) from which the slip rate of the reducing agent is ascertained is carried out via a combustion air ratio (.sub.v) upstream of the nitrogen oxide storage catalyst (3) and a combustion air ratio (.sub.n) downstream of the nitrogen oxide storage catalyst (3).

9. The method according to claim 1, wherein a selectivity expected value (E.sub.TS) is ascertained (90) from at least two of the expected values (E.sub.WPA, E.sub.BPU) for different models (20, 30) and this selectivity expected value (E.sub.TS) is compared to a threshold value (S.sub.TS) for the selectivity (120), wherein if the selectivity expected value (E.sub.TS) is greater than the threshold value (S.sub.TS) for the selectivity, the diagnosis (110) is classified as valid and if the selectivity expected value (E.sub.TS) is less than the threshold value (S.sub.TS) for the selectivity, the diagnosis (110) is classified as invalid (121) and is discarded.

10. A non-transitory computer-readable storage medium containing instructions that when executed by a computer cause the computer to control an exhaust system (2) of an internal combustion engine (1), in which a reduction of nitrogen oxides is carried out by means of a reducing agent, and performed during a regeneration of the nitrogen oxide storage catalyst (3), to: determine a slip rate (Q.sub.meas) of the reducing agent not absorbed in the nitrogen oxide storage catalyst (3); determine (61, 62) at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent from at least one model (20, 30); determine a monitoring variable (Q.sub.norm) based on the slip rate (Q.sub.meas) of the reducing agent ascertained from the measurement (10) and the at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent; and diagnose the storage capacity of the nitrogen oxide storage catalyst (3) based on the monitoring variable (Q.sub.norm).

11. An electronic computer device (6), which is configured to carry out monitoring of a nitrogen oxide storage catalyst (3) of an exhaust system (2) of an internal combustion engine (1), in which a reduction of nitrogen oxides is carried out by means of a reducing agent, and performed during a regeneration of the nitrogen oxide storage catalyst (3), by: determining a slip rate (Q.sub.meas) of the reducing agent not absorbed in the nitrogen oxide storage catalyst (3); determining (61, 62) at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent from at least one model (20, 30); determining a monitoring variable (Q.sub.norm) based on the slip rate (Q.sub.meas) of the reducing agent ascertained from the measurement (10) and the at least one expected value (E.sub.WPA, E.sub.BPU) for the slip rate of the reducing agent; and diagnosing the storage capacity of the nitrogen oxide storage catalyst (3) based on the monitoring variable (Q.sub.norm).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Exemplary embodiments of the invention are illustrated in the drawings and explained in greater detail in the following description.

[0031] FIG. 1 shows a schematic illustration of a nitrogen oxide storage catalyst in an exhaust system of an internal combustion engine, which can be monitored by means of an embodiment of the method according to the invention.

[0032] FIG. 2 shows a flow chart of an embodiment of the method according to the invention.

[0033] FIG. 3 shows a flow chart of the ascertainment of models as are used in the embodiment of the method according to the invention in FIG. 2.

DETAILED DESCRIPTION

[0034] FIG. 1 shows a schematic illustration of an internal combustion engine 1, an exhaust system 2, and an exhaust gas posttreatment unit having a nitrogen oxide storage catalyst 3, which can be monitored by means of one embodiment of the method according to the invention. An exhaust gas mass flow Q.sub.A from the internal combustion engine 1 is conducted via the exhaust system 2 to the exhaust gas posttreatment unit. The exhaust gas posttreatment unit can comprise, in addition to the nitrogen oxide storage catalyst 3, still further components for reducing pollutants, in particular nitrogen oxides.

[0035] An electronic control unit 6 controls the internal combustion engine 1, in particular its fuel injection (not shown separately). To save fuel, the internal combustion engine 1 is predominantly operated in a lean mode, in which more oxygen is present in the internal combustion engine 1 than is required for complete combustion of the fuel, so that an elevated concentration of nitrogen oxides is emitted by the internal combustion engine 1. The nitrogen oxides increasingly present in the exhaust gas in this mode are absorbed by the nitrogen oxide storage catalyst 3 and temporarily stored therein. To regenerate the nitrogen oxide storage catalyst 3, the internal combustion engine 1 is operated in a rich mode, in which less oxygen is present in the internal combustion engine 1 than is required for complete combustion of the fuel, so that hydrocarbons (HC), carbon monoxide (CO), and hydrogen (H.sub.2) are increasingly emitted. These reducing exhaust gas components are used as reducing agents for the nitrogen oxides in the nitrogen oxide storage catalyst 3 and reduce them to form nitrogen, which subsequently leaves the exhaust system 2.

[0036] In addition, a first lambda sensor 4 is arranged upstream of the nitrogen oxide storage catalyst 3, which measures the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and relays it to the electronic control unit 6. Furthermore, a second lambda sensor 5 is arranged downstream of the nitrogen oxide storage catalyst 3, which measures the combustion air ratio .sub.n downstream of the nitrogen oxide storage catalyst 3 and also relays it to the electronic control unit 6.

[0037] FIG. 2 shows a flow chart of an embodiment of the method according to the invention. During the regeneration of the nitrogen oxide storage catalyst 3, a measurement 10 of the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and of the combustion air ratio .sub.n downstream of the nitrogen oxide storage catalyst 3 is carried out and a measured slip rate Q.sub.meas is ascertained therefrom according to formula 1.

[00001]$\begin{array}{cc}{Q}_{\mathrm{meas}}=\frac{\left(1-\frac{1}{{}_n}\right)}{\left(1-\frac{1}{{}_v}\right)}& \left(\mathrm{Formula}.Math..Math.1\right)\end{array}$

[0038] In addition, a model 20 for the slip rate according to a WPA pattern (referred to in short as WPA model 20 hereafter) and a model 30 for the slip rate according to a BPU pattern (referred to as BPU model 30 in short hereafter) are provided. Reference is made to the embodiment of FIG. 3 for a more detailed description of the two models 20, 30. The following parameters are incorporated into each of the two models 20, 30, wherein the parameters for the WPA model 20 and the BPU model 30 differ:

[0039] The combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3; [0040] a setpoint value .sub.s of the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst; [0041] a deviation A.sub.vs between the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and the setpoint value .sub.s; [0042] the exhaust gas mass flow Q.sub.A; [0043] the temperature T.sub.cat of the nitrogen oxide storage catalyst 3; [0044] the running consumption m.sub.v of the reducing agent since regeneration start as a measure of the regeneration progress; [0045] the mass m.sub.catNOx of the nitrogen oxides stored in the nitrogen oxide storage catalyst 3; and/or [0046] the mass m.sub.catS of sulfur stored in the nitrogen oxide storage catalyst 3.

[0047] The running consumption m.sub.V of the reducing agent is computed as the integral of the product from the difference between the reciprocal combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and the reciprocal combustion air ratio .sub.n downstream of the nitrogen oxide storage catalyst 3 and the exhaust gas mass flow Q.sub.A according to formula 2. The integration start corresponds in this case to the time t.sub.0 from which the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 is less than 1.

[00002]$\begin{array}{cc}{m}_V={}_{t.Math..Math.0}^t.Math.\left(\frac{1}{{}_v}-\frac{1}{{}_n}\right).Math..Math.Q_A& \left(\mathrm{Formula}.Math..Math.2\right)\end{array}$

[0048] A modeled slip rate Q.sub.WPA according to the WPA pattern is ascertained from the WPA model 20 and a modeled slip rate Q.sub.BPU according to the BPU pattern is ascertained from the BPU model 30. Furthermore, an available reducing agent mass flow Q.sub.Ra is ascertained 40 from the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and the exhaust gas mass flow Q.sub.A. The ascertainment of the modeled slip rate Q.sub.WPA according to the WPA pattern and the ascertainment of the modeled slip rate Q.sub.BPU according to the BPU pattern and also the ascertainment 40 of the reducing agent mass flow Q.sub.Ra may be carried out simultaneously with one another and/or with the measurement 10 of the measured slip rate Q.sub.meas or in arbitrary sequence.

[0049] An expected value E.sub.WPA for the integral reducing agent slip mass of the WPA model 20 is ascertained by multiplying 51 the modeled slip rate Q.sub.WPA according to the WPA pattern with the available reducing agent mass flow Q.sub.Ra and subsequently integrating 61 the product over the measuring time t of the measurement 10. Similarly, an expected value E.sub.BPU for the BPU model 30 is ascertained by multiplying 52 the modeled slip rate Q.sub.BPU according to the BPU pattern with the available reducing agent mass flow Q.sub.Ra and subsequently integrating 62 the product over the measuring time of the measurement 10. Moreover, the available reducing agent mass flow Q.sub.Ra is also integrated 60 over the measuring time t of the measurement 10. By means of a division 71 of the expected value E.sub.WPA for the integral reducing agent slip mass of the WPA model 20 by the integrated available reducing agent mass flow Q.sub.Ra, a mean value Q.sub.WPA of the modeled slip rate Q.sub.WPA is obtained according to the WPA model 20 (referred to hereafter as mean value Q.sub.WPA for the WPA model). Similarly, by means of the division 72 of the expected value E.sub.BPU for the BPU model 20 by the integrated available reducing agent mass flow Q.sub.Ra, a mean value Q.sub.BPU of the modeled slip rate Q.sub.BPU is obtained according to the BPU model 20 (referred to hereafter as the mean value Q.sub.BPU for the BPU model). In the same manner, a mean value Q.sub.meas of the measured slip rate Q.sub.meas is obtained by multiplying 53 the measured slip rate Q.sub.meas with the available reducing agent mass flow Q.sub.Ra, subsequently integrating 63 over the measuring time t of the measurement 10, and finally dividing 73 by the integrated available reducing agent mass flow Q.sub.Ra. The integrations 60-63 are carried out simultaneously but can also be carried out in any arbitrary sequence.

[0050] Now the difference 80 between the mean value Q.sub.meas of the measured slip rate and the mean value Q.sub.WPA for the WPA model 20 is calculated, on the one hand, and the difference 90 between the mean value Q.sub.BPU for the BPU model and the mean value Q.sub.WPA for the WPA model is calculated, on the other hand, wherein the difference 90 between the mean value Q.sub.BPU for the BPU model and the mean value Q.sub.WPA for the WPA model represent a selectivity expected value E.sub.TS, the function of which will be explained hereafter. To finally obtain a monitoring variable, in this embodiment a scaled slip rate Q.sub.norm, which reflects the degree of damage of the nitrogen oxide storage catalyst 3, a division 100 is carried out of the difference 80 between the mean value Q.sub.meas of the measured slip rate and the mean value Q.sub.WPA for the WPA model 20 by the difference 90 between the mean value Q.sub.BPU for the BPU model and the mean value Q.sub.WPA for the WPA model according to following formula 3:

[00003]$\begin{array}{cc}{Q}_{\mathrm{norm}}=\frac{{\stackrel{\_}{Q}}_{\mathrm{meas}}-{\stackrel{\_}{Q}}_{\mathrm{WPA}}}{{\stackrel{\_}{Q}}_{\mathrm{BPU}}-{\stackrel{\_}{Q}}_{\mathrm{WPA}}}& \left(\mathrm{Formula}.Math..Math.3\right)\end{array}$

[0051] Finally, a diagnosis 110 is carried out on the basis of the scaled slip rate Q.sub.norm. For this purpose, it is checked which value range the scaled slip rate Q.sub.norm is in. If the scaled slip rate Q.sub.norm is in a first range less than zero, i.e., if the scaled slip rate Q.sub.norm assumes negative values, the state of the nitrogen oxide storage catalyst 3 is diagnosed as better than a state according to the WPA pattern, i.e., the nitrogen oxide storage catalyst is diagnosed 111 as completely intact. If the scaled slip rate Q.sub.norm is in a second range between zero and one, the state of the nitrogen oxide storage catalyst 3 is thus diagnosed as worse than the state according to the WPA pattern but better than the state according to the BPU pattern. Accordingly, there is damage of the nitrogen oxide storage catalyst 3, but this damage is still under the level relevant for the BPU pattern, and the nitrogen oxide storage catalyst is diagnosed 112 as sufficiently intact. If the scaled slip rate Q.sub.norm is in a third range greater than one, the state of the nitrogen oxide storage catalyst 3 is thus diagnosed as worse than the state according to the BPU pattern, i.e., the nitrogen oxide storage catalyst 3 is diagnosed 113 as defective.

[0052] In addition, the selectivity expected value E.sub.TS, which is a measure of the sensitivity in the present monitoring phase, is compared to a threshold value S.sub.TS. If the selectivity expected value E.sub.TS is greater than the threshold value S.sub.TS for the selectivity, the diagnosis 110 is thus classified as valid. If the selectivity expected value E.sub.TS is less than the threshold value S.sub.TS for the selectivity, however, the diagnosis 110 is thus classified 121 as invalid and the above-described result is discarded.

[0053] FIG. 3 shows a flow chart for ascertaining the models 20, 30 according to one embodiment of the method according to the invention. The ascertainment described hereafter can be applied similarly for the model 20 according to the WPA pattern and for the model 30 according to the BPU pattern, only the parameters described hereafter are selected and/or weighted differently in the two patterns. Characteristic maps or characteristic curves 210-215 are used to transfer the parameters into the modeled slip rate Q.sub.WPA or Q.sub.BPU. The characteristic curves 210-215 are assigned different data, whereby the parameters are weighted differently. At the beginning, a deviation A.sub.vs between the combustion air ratio .sub.v upstream of the nitrogen oxide storage catalyst 3 and the setpoint value .sub.s for the combustion air ratio .sub.v downstream of the nitrogen oxide storage catalyst 3 is ascertained by calculating a difference 200 between the two variables. A two-dimensional characteristic map 210 indicates a model base value for the slip rate as a function of the setpoint value .sub.s for the combustion air ratio and the deviation A.sub.vs between the combustion air ratio .sub.v and the setpoint value .sub.s. Separate one-dimensional characteristic curves 211-215 are applied in each case for the exhaust gas mass flow Q.sub.A, the temperature T.sub.cat of the nitrogen oxide storage catalyst 3, the present consumption m.sub.V of the reducing agent, the mass m.sub.catNOx of the nitrogen oxides and the mass m.sub.catS of sulfur, which are stored in the nitrogen oxide storage catalyst, and the obtained values are then multiplied 220 in the form of correction factors with the model base value for the slip rate, to obtain the modeled slip rate Q.sub.WPA or Q.sub.BPU, respectively, for the respective pattern.