Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SYSTEM AND METHOD FOR SELECTIVE CATALYTIC REDUCTION CONTROL
Document Type and Number:
WIPO Patent Application WO/2009/024622
Kind Code:
A2
Abstract:
A selective catalytic reduction (SCR) catalyst control system and method for an engine is disclosed. Urea injection to an SCR catalyst is determined based on an SCR catalyst model, which determines a value of stored NH3 in the SCR catalyst based on the NOx engine emission value, the SCR catalyst temperature, the quantity of urea supplied to the SCR catalyst and a pre-determined efficiency of conversion of NOx gases. A target value of stored NH3 and the value of stored NH3 in the SCR catalyst is then used to determine a stored NH3 differential, which is then used to calculate urea injection.

Inventors:
PARMENTIER MICHAEL (BE)
SCHMITT JULIEN (FR)
Application Number:
PCT/EP2008/063066
Publication Date:
February 26, 2009
Filing Date:
September 30, 2008
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
DELPHI TECH INC (US)
PARMENTIER MICHAEL (BE)
SCHMITT JULIEN (FR)
International Classes:
B01D53/94; B01D53/90; F01N3/20; F01N9/00; F01N11/00
Domestic Patent References:
WO2005068797A12005-07-28
Foreign References:
DE102005012568A12006-09-21
US20060000202A12006-01-05
US20070137181A12007-06-21
US20040098974A12004-05-27
Attorney, Agent or Firm:
MURGITROYD & COMPANY (165-169 Scotland Street, Glasgow Strathclyde G5 8PL, GB)
Download PDF:
Claims:

Claims

1. A selective catalytic reduction (SCR) catalyst control system (10) for an engine having an SCR catalyst comprising: a nitrogen oxides (NO x ) engine emission determination means (12) for determining a NO x engine emission value; a urea control means (20) enabled to supply a quantity of urea to the SCR catalyst; a NO x efficiency target means (18), which determines a stored ammonia (NH 3 ) target value in the SCR catalyst based on a target conversion NO x efficiency and an SCR catalyst temperature, determined from an SCR catalyst temperature determination means ; an SCR catalyst model (16) which determines a stored NH 3 value in the SCR catalyst based on the NO x engine emission value, the SCR catalyst temperature, the quantity of urea supplied to the SCR catalyst and a pre-determined efficiency of conversion of NO x gases; and chacterised by a NO x ratio calculation means (14) which determines a NO x ratio, being the ratio of nitrogen dioxide in the NO x engine emission value, and wherein the SCR catalyst model (16) takes into account the NO x ratio when determining the stored NH 3 value. differential determination means enabled to compare the stored NH 3 target value and the stored NH 3 value in the SCR catalyst to determine a stored NH 3 differential, wherein the urea control means determines the required quantity of urea to supply to the SCR catalyst based on the stored NH 3 differential.

2. A system (10) as claimed in claim 1 , wherein the NO x ratio calculation means (14) receives a first temperature value from a first temperature sensor and the NO x ratio is calculated according to the first temperature value.

3. A system (10) as claimed in claim 2, wherein the first temperature sensor measures oxidation catalyst temperature.

4. A system (10) as claimed in claim 2, wherein the first temperature sensor measures particulate filter temperature.

5. A system (10) as claimed in claim 2, wherein the first temperature sensor measures the first temperature value between a particulate filter and an oxidation catalyst.

6. A system (10) as claimed in any of claims 1 to 5, wherein the NO x engine emission determination (12) means is an engine-out NO x model.

7. A system (16) as claimed in claim 6, wherein the engine-out NO x model (12) calculates the NO x engine emission value based on injected fuel flow into the engine, engine load, Exhaust Gas Recirculation (EGR) rate and ambient temperature.

8. A system (10) as claimed in any of claims 1 to 5, wherein the NO x engine emission determination (12) means is a NO x sensor positioned upstream of the SCR catalyst, which provides a NO x engine emission value.

9. A system (10) as claimed in any of claims 1 to 8, wherein the NO x efficiency target means (18) further utilises values from sensors measuring one or more of the following parameters to determine a target value of stored ammonia (NH 3 ) in the SCR catalyst: engine speed; engine load; air temperature; coolant temperature; or Diesel Particulate Filter (DPF) regeneration mode.

10. A system as claimed in any of claims 1 to 9, wherein the SCR model (16) determines a NH 3 slip value, which represents the amount of NH 3 exiting the SCR catalyst, by calculating an SCR catalyst capacity based on physical characteristics of the SCR catalyst and the SCR catalyst temperature and taking into account the value of stored NH 3 in the SCR catalyst.

11. A system as claimed in any of claims 1 to 10, further comprising NH 3 slip control means and a NO x engine emission increasing means, wherein, if it is determined that the NH 3 slip value is above, or is predicted to rise above, a pre-determined value, NO x engine emission increasing means is directed to increase NO x engine emissions, thereby reducing NH 3 slip.

12. A system as claimed in claim 11 , wherein the NO x engine emission increasing means is a Exhaust Gas Recirculation (EGR) means, wherein NO x engine emissions are increased by reducing or stopping the amount of EGR to the engine.

13. A system as claimed in any of claims 1 to 12, wherein the system further comprises an SCR model modification means (40, 42, 46, 48).

14. A system as claimed in claim 13, wherein the SCR model modification means comprises a NH 3 sensor (42) enabled to measure actual NH 3 slip from the SCR catalyst, actual NH 3 slip averaging means and SCR model NH 3 slip value averaging means, and comparison means enabled to compare outputs from the actual NH 3 slip averaging means and the SCR model NH 3 slip value averaging means and determine an NH 3 slip estimation error (46, 48), wherein the SCR model is modified by the SCR model modification means according to the NH 3 slip estimation error,

15. A system as claimed in claims 13 or 14, wherein the SCR model modification means is enabled to modify the SCR model by altering the pre-determined efficiency of conversion of NO x gases based on the NH 3 slip estimation error (46, 48).

16. A system as claimed in any of claims 13 to 15, wherein the SCR model modification means is enabled to modify the SCR model by altering the SCR catalyst capacity (46) based on the NH 3 slip estimation error.

17. A system as claimed in claim 16, wherein the SCR model modification means modifies the SCR model by altering the SCR catalyst capacity if the SCR catalyst is filled with NH 3 by a pre-determined minimum amount for a predetermined time.

18. A method of controlling selective catalytic reduction (SCR) in an engine having an SCR catalyst comprising the steps of:

(i) determining a nitrogen oxides (NO x ) engine emission value; (ii) controlling supply of a quantity of urea to the SCR catalyst; (iii) measuring an SCR catalyst temperature from the SCR catalyst

(iv) determining a stored ammonia (NH 3 ) target value in the SCR catalyst based on a target NO x conversion efficiency and the SCR catalyst temperature;

(v) calculating a NO x ratio, being the ratio of nitrogen dioxide in the NO x engine emission value;

(vi) calculating a stored NH 3 value in the SCR catalyst using an SCR catalyst model based on the NO x engine emission value, the SCR catalyst temperature, the quantity of urea supplied to the SCR catalyst, the NOx ratio and a pre-determined efficiency of conversion of NO x gases; and

(vii) comparing the target value of stored NH 3 and the value of stored NH 3 in the SCR catalyst to determine a stored NH 3 differential, wherein step (ii) controls the supply of the required quantity of urea to the SCR catalyst based on the stored NH 3 differential.

19. A method as claimed in claim 18, wherein the step of calculating a NO x ratio comprises measuring a first temperature value from a first temperature sensor and calculating the NO x ratio according to the first temperature value.

20. A method as claimed in claim 19, wherein the first temperature value is an oxidation catalyst temperature value.

21. A method as claimed in claim 19, wherein the first temperature value is a particulate filter temperature.

22. A method as claimed in claim 19, wherein the first temperature value is measured between a particulate filter and an oxidation catalyst.

23. A method as claimed in any of claims 18 to 22, wherein step (i) comprises calculating the (NO x ) engine emission value based on an engine-out NO x model.

24. A method as claimed in claim 23, wherein the step of calculating the (NO x ) engine emission value takes injected fuel flow into the engine, engine load, Exhaust Gas Recirculation (EGR) rate and ambient temperature into account in the engine-out NO x model.

25. A method as claimed in any of claims 18 to 22, wherein step (i) comprises measuring a NO x engine emission value from a NO x sensor positioned upstream of the SCR catalyst.

26. A method as claimed in any of claims 18 to 25, wherein step (iv) further comprises measuring one or more of the following parameters to determine a target value of stored ammonia (NH 3 ) in the SCR catalyst: engine speed; engine load; air temperature; coolant temperature; or Diesel Particulate Filter (DPF) regeneration mode.

27. A method as claimed in any of claims 18 to 26, wherein step (v) further comprises determining an NH 3 slip value, which represents the amount of NH 3 exiting the SCR catalyst, by calculating, within the SCR model, an SCR catalyst capacity based on physical characteristics of the SCR catalyst and the SCR catalyst temperature and taking into account the value of stored NH 3 in the SCR catalyst.

28. A method as claimed in any of claims 18 to 27, further comprising the steps of controlling NH 3 slip and increasing NO x engine emissions, wherein, if it is determined that the NH 3 slip value is above, or is predicted to rise above, a pre-determined value, NO x engine emissions are increased, thereby reducing NH 3 slip.

29. A method as claimed in claim 28, wherein the step of increasing NO x engine emissions comprises reducing or stopping the amount of Exhaust

Gas Recirculation (EGR) to the engine.

30. A method as claimed in any of claims 18 to 29, further comprising the step of modifying the SCR model used in step (vi).

31. A method as claimed in claim 30, wherein the step of modifying the SCR model comprises measuring actual NH 3 slip from the SCR catalyst using an NH 3 sensor, calculating the SCR model NH 3 slip from the SCR model, averaging the actual NH 3 slip over a pre-determined time and averaging SCR model NH 3 slip over the same pre-determined time, comparing the averaged actual NH 3 slip and the averaged SCR model NH 3 slip and determining an NH 3 slip estimation error, wherein the SCR model is subsequently modified according to the NH 3 slip estimation error.

32. A method as claimed in claim 31 , wherein the step of modifying the SCR model modifies the SCR model by altering the pre-determined efficiency of conversion of NO x gases based on the NH 3 slip estimation error.

33. A method as claimed in claim 30 or claim 31 , wherein the step of modifying the SCR model modifies the SCR model by altering the SCR catalyst capacity based on the NH 3 slip estimation error.

34. A method as claimed in claim 33, wherein the step of modifying the SCR model modifies the SCR model by altering the SCR catalyst capacity if the SCR catalyst is filled with NH 3 by a pre-determined minimum amount for a predetermined time.

35. A Diesel engine incorporating a selective catalytic reduction (SCR) catalyst control system according claims 1 to 17.

Description:

System and method for selective catalytic reduction control

The present invention relates to control of Selective Catalytic Reduction (SCR) catalyst and particularly, but not exclusively, to control of an SCR catalyst in vehicle engines.

Selective Catalytic Reduction (SCR) catalysts remove nitrogen oxides (NO x ), often the most abundant and polluting component in exhaust gases, through a chemical reaction between the exhaust gases, a reducing agent, and a catalyst.

The control of Selective Catalytic Reduction (SCR) catalysts consists of injecting an amount of reducing agent, typically urea, also known as carbamide ((NH 2 )2CO), which decomposes to ammonia (NH3) and carbon dioxide in the presence of water, oxygen and heat. Ammonia then reacts with NO x gases to produce nitrogen and water. The amount of reducing agent injected is required to provide the maximum NO x conversion efficiency whilst keeping excess NH 3 , also known as NH3 slip, to low values. SCR catalysts have mainly been introduced on heavy duty vehicles where high NO x levels are present and where steady state can be considered to be the main operating conditions. In these conditions, SCR control consists of supplying a certain NH 3 to NO x ratio, usually mapped as a function of speed and load.

Applying this kind of control on a passenger car, where transient conditions are more frequent, usually requires specific transient corrections. Moreover, vanadium based catalysts are often used on heavy duty vehicles and this technology is known to have a reduced buffering effect (the temporary storage of NH 3 ) than new Zeolite based catalysts (Fe, Cu) used on passenger car (or light duty) applications.

Another possible way to control an SCR catalyst is to model the chemical behaviour of the catalyst and implement the model in the Engine Control Unit (ECU). This approach requires a high amount of calibration work in order to identify all the parameters that need to be taken into account in a chemical model. The calculation load required for this approach is also very high as it would require to calculate multiple complex chemical reactions occurring in the catalyst in slices along its length. Chemical reactions in the catalyst depend on the temperature of the catalyst and on the concentration of the different compounds, which vary along the length of the catalyst, especially during transients. So to obtain an accurate model of all the reactions in the catalyst, the calculations would require to model several slices in series. Closed-loop control of such a system is usually realised with NO x sensors before and after the SCR catalyst, but NO x sensors are also sensitive to NH 3 creating additional difficulty that is required to be taken into account by the closed-loop control.

DE 102005012568 discloses a device and method for removing nitrogen oxide from the exhaust of an internal combustion engine. An aggregate containing a reducing agent is added to the exhaust dependent on variables such as engine load, air/fuel ratio and engine revolutions.

According to a first aspect of the present invention there is provided a selective catalytic reduction (SCR) catalyst control system for an engine having an SCR catalyst comprising: a nitrogen oxides (NO x ) engine emission determination means for determining a NO x engine emission value; a urea control means enabled to supply a quantity of urea to the SCR catalyst;

a NO x efficiency target means, which determines a target value of stored ammonia (NH 3 ) in the SCR catalyst based on a required NO x efficiency and an SCR catalyst temperature determined from an SCR catalyst temperature determination means ; an SCR catalyst model which determines a value of stored NH 3 in the SCR catalyst based on the NO x engine emission value, the SCR catalyst temperature, the quantity of urea supplied to the SCR catalyst and a pre-determined efficiency of conversion of NO x gases; and differential determination means enabled to compare the target value of stored NH 3 and the value of stored NH 3 in the SCR catalyst to determine a stored NH 3 differential, wherein the urea control means determines the required quantity of urea to supply to the SCR catalyst based on the stored NH 3 differential.

Preferably, the NO x ratio calculation means receives a first temperature value from a first temperature sensor and the NO x ratio is calculated according to the first temperature value.

Preferably, the first temperature sensor measures oxidation catalyst temperature.

Alternatively, the first temperature sensor measures particulate filter temperature.

Alternatively, the first temperature sensor measures the first temperature value between a particulate filter and an oxidation catalyst

Preferably, the NO x engine emission determination means is an engine- out NO x model.

Preferably, the engine-out NO x model calculates the NO x engine emission value based on injected fuel flow into the engine, engine load, Exhaust Gas Recirculation (EGR) rate and ambient temperature.

Alternatively, the NO x engine emission determination means is a NO x sensor positioned upstream of the SCR catalyst, which provides a NO x engine emission value.

Preferably, the NO x efficiency target means further utilises values from sensors measuring one or more of the following parameters to determine a target value of stored ammonia (NH 3 ) in the SCR catalyst: engine speed; engine load; air temperature; coolant temperature; or Diesel Particulate Filter (DPF) regeneration mode.

Preferably, the SCR model determines a NH 3 slip value, which represents the amount of NH 3 exiting the SCR catalyst, by calculating an SCR catalyst capacity based on physical characteristics of the SCR catalyst and the SCR catalyst temperature and taking into account the value of stored NH 3 in the SCR catalyst.

Preferably, the system further comprises NH 3 slip control means and a NO x engine emission increasing means, wherein, if it is determined that the NH 3 slip value is above, or is predicted to rise above, a pre-determined value, NO x engine emission increasing is directed to increase NO x engine emissions, thereby reducing NH3 slip.

Preferably, the NO x engine emission increasing means is a Exhaust Gas Recirculation (EGR) means, wherein NO x engine emissions are increased by reducing or stopping the amount of EGR to the engine.

Preferably, the system further comprises an SCR model modification means.

Preferably, the SCR model modification means comprises a NH 3 sensor enabled to measure actual NH3 slip from the SCR catalyst, actual NH 3 slip averaging means and SCR model NH 3 slip value averaging means, and comparison means enabled to compare outputs from the actual NH 3 slip averaging means and the SCR model NH 3 slip value averaging means and determine an NH 3 slip estimation error, wherein the SCR model is modified by the SCR model modification means according to the NH 3 slip estimation error.

Preferably, the SCR model modification means is enabled to modify the SCR model by altering the pre-determined efficiency of conversion of NO x gases based on the NH 3 slip estimation error.

Alternatively or further preferably, the SCR model modification means is enabled to modify the SCR model by altering the SCR catalyst capacity based on the NH 3 slip estimation error.

Preferably, the SCR model modification means modifies the SCR model by altering the SCR catalyst capacity if the SCR catalyst is filled with NH 3 by a pre-determined minimum amount for a predetermined time.

According to a second aspect of the present invention there is provided a method of controlling selective catalytic reduction (SCR) in an engine having an SCR catalyst comprising the steps of:

(i) determining a nitrogen oxides (NO x ) engine emission value; (ii) controlling supply of a quantity of urea to the SCR catalyst; (iii) measuring an SCR catalyst temperature from the SCR catalyst

(iv) determining a stored ammonia (NH 3 ) target value in the SCR catalyst based on a target NO x conversion efficiency and the SCR catalyst temperature;

(v) calculating a NO x ratio, being the ratio of nitrogen dioxide in the NO x engine emission value;

(vi) calculating a stored NH 3 value in the SCR catalyst using an SCR catalyst model based on the NO x engine emission value, the SCR catalyst temperature, the quantity of urea supplied to the SCR catalyst, the NOx ratio and a pre-determined efficiency of conversion of NO x gases; and

(vii) comparing the target value of stored NH 3 and the value of stored NH 3 in the SCR catalyst to determine a stored NH 3 differential, wherein step (ii) controls the supply of the required quantity of urea to the SCR catalyst based on the stored NH 3 differential.

Preferably the step of calculating a NO x ratio comprises measuring a first temperature value from a first temperature sensor and calculating the NO x ratio according to the first temperature value.

Preferably, the first temperature value is an oxidation catalyst temperature value.

Alternatively, the first temperature value is a particulate filter temperature.

Further alternatively, the first temperature value is measured between a particulate filter and an oxidation catalyst.

Preferably, step (i) comprises calculating the (NO x ) engine emission value based on an engine-out NO x model.

Preferably, the step of calculating the (NO x ) engine emission value takes injected fuel flow into the engine, engine load, Exhaust Gas Recirculation (EGR) rate and ambient temperature into account in the engine-out NO x model.

Alternatively, step (i) comprises measuring a NO x engine emission value from a NO x sensor positioned upstream of the SCR catalyst.

Preferably, step (Iv) further comprises measuring one or more of the following parameters to determine a target value of stored ammonia (NH 3 ) in the SCR catalyst: engine speed; engine load; air temperature; coolant temperature; or Diesel Particulate Filter (DPF) regeneration mode.

Preferably, step (vi) further comprises determining a NH 3 slip value, which represents the amount of NH 3 exiting the SCR catalyst, by calculating, within the SCR model, an SCR catalyst capacity based on physical characteristics of the SCR catalyst and the SCR catalyst temperature and taking into account the value of stored NH 3 in the SCR catalyst.

Preferably, the method further comprises the steps of controlling NH 3 slip and increasing NO x engine emissions, wherein, if it is determined that the NH 3 slip value is above, or is predicted to rise above, a pre-determined value, NO x engine emissions are increased, thereby reducing NH 3 slip.

Preferably, the step of increasing NO x engine emissions comprises reducing or stopping the amount of Exhaust Gas Recirculation (EGR) to the engine.

Preferably, the method further comprises modifying the SCR model used in step (vi).

Preferably, the step of modifying the SCR model comprises measuring actual NH 3 slip from the SCR catalyst using an NH 3 sensor, calculating the SCR model NH 3 slip from the SCR model, averaging the actual NH 3 slip over a pre-determined time and averaging SCR model NH 3 slip over the same pre-determined time, comparing the averaged actual NH 3 slip and the averaged SCR model NH 3 slip and determining an NH 3 slip estimation error, wherein the SCR model is subsequently modified according to the NH 3 slip estimation error.

Preferably, the step of modifying the SCR model modifies the SCR model by altering the pre-determined efficiency of conversion of NO x gases based on the NH 3 slip estimation error.

Alternatively or further preferably, the step of modifying the SCR model modifies the SCR model by altering the SCR catalyst capacity based on the NH 3 slip estimation error.

Preferably, the step of modifying the SCR model modifies the SCR model by altering the SCR catalyst capacity if the SCR catalyst is filled with NH 3 by a pre-determined minimum amount for a predetermined time.

According to a third aspect of the present invention there is provided a Diesel engine incorporating a selective catalytic reduction (SCR) catalyst control system according to the first aspect of the present invention.

Embodiments of the present invention will now be described, by way of example only, with reference to the drawings, in which:

Fig. 1 is a graph of measured and modelled NO x emissions;

Fig. 2 shows a flow diagram of an open loop control structure for a Selective Catalytic Reduction (SCR) catalyst;

Fig. 3 is a graph showing NO x efficiency against stored NHb;

Fig. 4 is a graph showing variations in NH 3 slip depending on control of NO x emissions;

Fig. 5 is a flow diagram of an open loop control structure for a Selective Catalytic Reduction catalyst with an additional closed loop control section; and

Fig. 6 is a graph showing a temperature transient showing modelled NH3 outflow from the catalyst, measures NH3 concentration from a sensor and temperature of gases at the SCR inlet.

As explained above, the engine-out NO x flow needs to be known to inject the correct amount of urea, as too little gives poor efficiency and too much gives NH3 slip. Prior art systems use a NO x sensor located upstream of the SCR catalyst.

The present invention uses simplified models to calculate the NO x conversion efficiency of the SCR catalyst and drive the injection quantity of urea required to maintain this efficiency to a certain level. The model can take into account several key parameters including:

total NO x flow entering the SCR catalyst and NO 2 /NO X ratio; SCR catalyst temperature; and stored NH3 in the SCR catalyst.

The total NO x flow can be measured by a NO x sensor located before the SCR.

Alternatively, as used in a preferable embodiment on the present invention, a NO x model can be used that replaces the NO x sensor. The NO x flow is modelled as a fraction of injected fuel flow. This fraction is mapped as a function of engine load (IMEP - Indicated Mean Effective Pressure), corrected for inert EGR (Exhaust Gas Recirculation) rate and ambient temperature. Fig. 1 shows the comparison between modelled and measured NO x using a US city cycle for passenger cars called FTP75 or EPA III.

Referring now to Fig. 2, an open loop model for selective catalytic reduction control 10 in an exhaust system of an engine comprises an engine-out NO x model 12, a Nθ2/NO x ratio model 14, an SCR model 16, which models storage of NH 3 and NO x conversion efficiency, a NO x efficiency target model 18 and an urea injection control 20.

The engine-out NO x model 12, as described above, uses injected fuel flow to calculate the NO x emission taking into account engine load, EGR rate and ambient temperature.

The Nθ 2 /NO x ratio model 14, calculates a Nθ 2 /NO x ratio based on a temperature measurement from an oxidation catalyst in the exhaust system. Although, the temperature measurement may come from an alternative position in the exhaust system. In particular, the temperature

could be measured between the oxidation catalyst and a particulate filter or even after the particulate filter. Furthermore, a particulate filter can effect the NO 2 /NO X ratio and therefore can be taken into consideration in the NO 2 /NOχ ratio model 14.

The SCR model 16 calculates the NO x conversion efficiency of an SCR in the exhaust system based on a function of stored NH 3 in the SCR, the amount of injected urea from the urea injection control 20, and temperature. This efficiency can then be corrected for the N(VNO x ratio obtained from the NO x ratio model 14. When the NO x conversion efficiency is known, the amount of NH3 used for NO x reduction can be calculated based on a pre-determined amount of NH 3 being required to reduce a predetermined amount of NO x and, as such the amount of stored NH 3 is calculated, along with any excess NH 3 (NH 3 slip) or any output of NO x gases. Since the storage capacity of an SCR catalyst decreases with temperature, if the SCR temperature increases too quickly, a quantity of stored NH 3 will be released. The maximum storage capacity for NH 3 of the SCR catalyst is calculated as a function of SCR temperature. Therefore, NH 3 leaving the SCR (NH 3 slip) is also an output of the SCR model 16. This provides a means to compare the SCR model 16 to a NH 3 sensor located downstream of the SCR catalyst (discussed in more detail below with respect to closed loop control).

The NO x efficiency target model 18 generates a target value for stored NH 3 , NO x conversion efficiency being dependent on stored NH 3 in the SCR catalyst, based on the required NO x efficiency and the SCR temperature. The target NO x efficiency is then corrected for other conditions including engine speed, engine load, air temperature, coolant temperature, and Diesel Particulate Filter (DPF) regeneration mode, if

appropriate to the system. An example of NO x efficiency with respect to stored NH 3 is shown in Fig. 3.

The target stored NH3 value from the NO x efficiency target model 18 is then compared to the calculated stored NH 3 value to generate a stored NH3 differential 22, which is the difference between what stored NH 3 currently is and where it requires to be (target stored NH 3 ).

The urea injection control 20 has, as its inputs, the stored NH3 differential 22 and an exhaust gas temperature value, measured directly from the exhaust gas. Based on the NH 3 differential 22 a urea injection amount is calculated, in this case, using a proportional gain controller, to bring the calculated stored NH 3 value to the target stored NH 3 value. To account for oxidation of NH 3 at high temperature and/or lack of urea hydrolysis at low temperature, the amount of injected urea is modified by a map function of exhaust gas temperature.

When NH 3 slip occurs, the amount of NH 3 leaving the SCR can be reduced by increasing the rate of NH 3 consumption in the SCR. To achieve this, an increase in NO x flow in the SCR is required, as the NH3 reacts with NO x . In this example, and as shown in Fig. 4, an increase in NO x flow is achieved by turning off EGR. SCR temperature 30 is shown increasing around 200 0 C to around 300 0 C. If no action is taken, that is EGR is left as normal, NO x emissions 32, from the engine, stays at around 40 ppm (parts per million) but NH 3 slip 34 increases dramatically to more than 100ppm, which is the saturation of the sensor used. Conversely, if EGR is turned off, NO x emissions 36 from the engine increase to around IOOppm but NH 3 slip 38 peaks at about 70ppm before reducing. It should be noted that NO x emissions 32, 36 are from the engine and not vented from the exhaust. NH 3 slip is caused by having too much NH 3 in the SCR

catalyst with respect to NO x and, therefore, the NO x conversion efficiency will be as high as is possible.

Additionally, the urea injection can be shut-off to reduce the amount of NH 3 in the SCR catalyst. These actions are taken when the NH 3 slip, either computed by the SCR model 16 or measured by a NH 3 sensor, goes above a pre-determined threshold.

As indicated above, an important factor of the SCR model 16 is the stored NH 3 mass. This calculation of the amount of stored NH 3 can be, or certainly become, inaccurate for a number of reasons during implementation in a real world system. For example, as the SCR catalyst ages, its capacity and efficiency can decrease over time causing an error in the estimation of stored NH3. Furthermore, if the engine produces a different level of NO x than modelled due to engine to engine variations or aging of the engine or the urea flow is different than expected, again, an error in the estimation of stored NH 3 can occur.

An accurate physical modelling of the NH3 slip of the catalyst is too complicated to be implemented and calibrated in an ECU (Engine Control Unit), as it would involve a complex chemical model. The SCR mode! 16 is an average estimator of the NH 3 slip. The transient behaviour of the SCR model 16 is not accurate because an SCR catalyst's capacity for storing NH3 is directly linked to a temperature which is supposed to be constant over the length of the SCR catalyst. But the average NH 3 mass predicted by the SCR model 16 can be compared to the average NH3 mass seen by a NH3 sensor downstream of the SCR catalyst in the exhaust system.

Accordingly, referring to Fig. 5, a NH3 slip model 40 and a NH3 slip sensor 42 compare the NH 3 slip from the mode) and the sensor respectively. A

NH 3 slip error 44 is then generated before being passes to a capacity error 46 means and an efficiency error means 48.

Under speciai conditions, where the model is known to have the best precision, including a high enough DOC temperature and SCR temperature within a predetermined range, the NH3 slip out of the SCR model 16 and the NH 3 flow seen by the NH 3 sensor are monitored over a fixed time. Once the time has elapsed, both values should be equal if the SCR model 16 is accurate.

Referring to Fig. 6, SCR inlet temperature 50 is shown increasing from around 200C to 350C during a temperature transient. NH 3 slip modelled value 52 and NH 3 slip sensor value 54 both show an increase in NH3 slip as a result, although there is time lag associated with the NH 3 slip sensor value 54.

If the monitoring period is long enough, transient errors can be neglected and after a pre-determiπed time, if an error remains between the modelled and sensed NH 3 slip, the SCR model 16 can be altered accordingly.

According to various factors associated with the SCR model 16 the capacity of the SCR catalyst has to be corrected or the efficiency of the NO x conversion modified. A capacity correction is made if the SCR catalyst is filled with NH3 by a pre-determined minimum amount for a predetermined time. Otherwise, it performs a modification to the efficiency of the NO x conversion. The modification to the efficiency of the NO x conversion can also correct the engine-out NO x model in case this input parameter is wrong. A modification to the pre-determined efficiency of the NO x conversion acts as a global modifier as it corrects SCR efficiency, injector error, NO x flow model error and urea quality. As the target capacity

of the SCR catalyst may be set quite low to avoid IMH3 slip during normal operation, the conditions required for a capacity correction may never occur. As such, it can be beneficial to increase the NH 3 stored target so that a capacity correction can occur ( and the SCR catalyst is filled to its maximum capacity). An efficiency modification occurs more often because, as mentioned previously, certain conditions are required for a capacity correction.

Further modifications and improvements may be made without departing from the scope of the present invention.