Method for estimating the degradation of the trapping capacity of NOx-Trap type catalytic converter

Abstract

A method for estimating the degradation of the trapping capacity of a NOx-Trap catalytic converter, in accordance with which, if the actual duration of a first NOx regeneration process is equal to the predetermined duration, it is assumed that the trapping capacity is unchanged; if the actual duration of the first NOx regeneration process is less than the predetermined duration, at least one corrective action is performed in order to attempt to counteract the degeneration of the NOx-Trap catalytic converter, a subsequent NOx regeneration process is performed, if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then new characteristic operating parameters for the corrective action are used for the subsequent life of the NOx-Trap catalytic converter whereas, if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, the estimated trapping capacity of the NOx-Trap catalytic converter is reduced.

Description

PRIORITY CLAIM

[0001] This application claims priority from Italian patent application No. BO2003A filed on Mar. 13, 2003, which is incorporated herein by reference.

TECHNICAL FILED

[0002] The present invention relates to a method for estimating the degradation of the trapping capacity of a NOx-Trap type catalytic converter.

[0003] The present invention is advantageously applied to an internal combustion automotive engine supplied with fuel by direct injection of petrol into the cylinders and having combustion with a lean mixture and stratified charge, to which the following description will make explicit reference without thereby restricting the general scope.

BACKGROUND

[0004] A direct injection petrol engine comprises an exhaust manifold, which communicates with the cylinders by means of the respective exhaust valves and terminates in an exhaust pipe equipped with a precatalytic converter capable of promoting the conversion of the NO groups produced during combustion into NO2 and a subsequent NOx-Trap catalytic converter capable of trapping the NOx groups and so preventing their release into the atmosphere. The NOx-Trap catalytic converter traps within itself both the NOx groups produced during combustion and the sulfur (in the form of SOx) that is contained in the fuel and is released during combustion; moreover, the NOx-Trap catalytic converter itself has a limited trapping capacity (generally of between 3 and 5 grams) and when said trapping capacity is exhausted, the NOx-Trap catalytic converter must be cleaned by a regeneration process.

[0005] The total mass of NOx groups produced during combustion is much greater than the mass of sulfur released during combustion; moreover the regeneration process to remove NOx groups (a few seconds of rich combustion) is much shorter than the regeneration process to remove sulfur (of the order of 30-60 seconds of rich combustion combined with an internal temperature in the catalytic converter that is very much higher than the normal operating temperature). For the reasons stated above, the regeneration process to remove NOx groups is normally carried out every 45-75 seconds of engine operation, whereas the regeneration process to remove sulfur is normally carried out every 6-12 hours of engine operation.

[0006] The regeneration processes are scheduled by a central control unit using a storage model for the NOx-Trap catalytic converter, said model being based on a knowledge of the estimated trapping capacity of the NOx-Trap catalytic converter, and using a model of the production of NOx and SOx group by the engine. Each time NOx regeneration of the NOx-Trap catalytic converter is performed, the central control unit checks, using the signal from a lambda probe and/or a NOx probe provided downstream from the NOx-Trap catalytic converter, whether the actual duration of the NOx regeneration process is less than a predetermined value on the basis of the current model; if this is the case, i.e. if the actual duration of the NOx regeneration process is less than the predetermined value, it is clear that the NOx-Trap catalytic converter has trapped a smaller quantity of NOx than predicted and the central control unit accordingly assumes that said phenomenon is due to degradation of the NOx-Trap catalytic converter and reduces the estimate of the trapping capacity of the NOx-Trap catalytic converter used in the storage model for the NOx-Trap catalytic converter.

[0007] However, experimental trials have revealed that, when the above-described method is used, there is a tendency to underestimate the actual trapping capacity of the NOx-Trap catalytic converter, with a consequent increase in fuel consumption (and thus in the level of atmospheric emissions), due to the fact that underestimating the actual trapping capacity of the NOx-Trap catalytic converter results in NOx regeneration processes being carried out more frequently.

SUMMARY

[0008] The object of the present invention is to provide a method for estimating the degradation of the trapping capacity of a NOx-Trap type catalytic converter, which method does not have the above-described disadvantages and, in particular, is simple and economical to implement.

[0009] The present invention provides a method for estimating the degradation of the trapping capacity of a NOx-Trap type catalytic converter as defined in claim 1 and, preferably, in any one of the subsequent claims directly or indirectly subordinate to claim 1.

BRIEF DESCRIPTION OF THE DRAWING

[0010] The present invention will now be described with reference to the attached drawing, which illustrates a non-limiting embodiment; in particular, the attached figure is a schematic view of an internal combustion engine, which is controlled by a control unit that implements the estimation method provided by the present invention.

DETAILED DESCRIPTION

[0011] In the attached FIG. 1 denotes the overall internal combustion engine equipped with four cylinders 2 (only one of which is shown in FIG. 1), each of which is connected to an intake manifold 3 via at least one respective intake valve 4 and to an exhaust manifold 5 via at least one respective exhaust valve 6. The intake manifold 3 receives fresh air (i.e., air originating from the outside environment) via a throttle valve 7 that is adjustable between a closed position and a maximally open position. Petrol is injected directly into each cylinder 2 by a respective injector 8.

[0012] From the exhaust manifold 5 there leaves an exhaust pipe 9, which comprises a catalytic preconverter 10 and a subsequent NOx-Trap catalytic converter 11; inside the exhaust pipe 9 there is installed a UEGO probe 12, which is arranged upstream from the catalytic preconverter 10 and is capable of detecting the quantity of oxygen present in the exhaust gases entering the catalytic preconverter 10, a temperature sensor 13, which is arranged between the catalytic preconverter 10 and NOx-Trap catalytic converter 11 and is capable of detecting the temperature of the gases entering the NOx-Trap catalytic converter 11, and a multisensor 14, which is arranged downstream from the NOx-Trap catalytic converter 11 and is capable of detecting either the presence of NOx groups (nitrogenous group sensor) or the quantity of oxygen present relative to stoichiometric conditions (lambda probe) in the exhaust gases leaving the NOx-Trap catalytic converter 11 (i.e., in the exhaust gases released from the exhaust pipe 9 into the atmosphere).

[0013] The engine 1 furthermore comprises a control unit 15 which, inter alia, on each cycle controls the throttle valve 7 and the injector 8 to fill the cylinder 2 with a quantity of combustion agent (fresh air) and fuel in a specific ratio as a function of the operating conditions of the engine 1 and as a function of the commands received from the driver; in particular, the control unit 15 is capable of causing the engine 1 to operate by combustion with a lean mixture and stratified charge. In order to allow the control unit 15 to capture the data required for correct operation thereof, the control unit 15 is connected to the UEGO probe 12, the temperature sensor 13 and the multisensor 14.

[0014] In service, the NOx-Trap catalytic converter 11 stores either the NOx groups produced during combustion or the sulfur (in the form of SOx) contained in the fuel and released during combustion, in order to prevent said constituents from being released directly into the atmosphere. The NOx-Trap catalytic converter 11 has a limited trapping capacity C for NOx groups and sulfur (normally amounting to 4 grams) and when said trapping capacity C is exhausted the NOx-Trap catalytic converter 11 must be cleaned by means of a regeneration process. The total mass of NOx groups produced during combustion is much greater than the mass of sulfur released during combustion, and moreover the regeneration process to remove NOx groups (a few seconds of rich combustion of the engine) is much shorter than the regeneration process to remove sulfur (indicatively 30-60 seconds of rich combustion combined with an internal temperature in the NOx-Trap catalytic converter 11 that is very much higher than the normal operating temperature). For the reasons stated above, the regeneration process to remove NOx groups is normally carried out every 45-75 seconds of operation of the engine 1, whereas the regeneration process to remove sulfur (also known as the desulfation process) is normally carried out every 6-12 hours of operation of the engine 1.

[0015] The regeneration processes are scheduled by the central control unit 15 using a storage model for the NOx-Trap catalytic converter 11, said model being stored in a memory 16 and based on a knowledge of the estimated trapping capacity C of the NOx-Trap catalytic converter 11, and using a model of the production of NOx and SOx groups by the engine 1, said model being stored in the memory 16. In particular, the quantity of NOx produced by the engine 1 is obtained in known manner by the control unit 15 using maps that state the specific quantity (i.e., the quantity per unit of fuel injected into the cylinders 2) of NOx and SOx groups produced by the engine 1 as a function of engine status (typically as a function of engine speed and as a function of delivered torque). As is known, the above-mentioned models are determined by means of a theoretical analysis of the systems and by means of a series of laboratory tests carried out on the engine 1 equipped with a series of auxiliary measurement sensors, which are capable of providing an instantaneous and accurate measurement of all the parameters involved in the operation of the engine 1 itself.

[0016] During normal operation of the engine 1 and using the storage model for the NOx-Trap catalytic converter 11, the control unit 15 estimates the quantity of NOx groups stored in the NOx-Trap catalytic converter 11; when said quantity of stored NOx groups exceeds a predetermined threshold, the control unit 15 triggers performance of the NOx regeneration process. The NOx regeneration process is of a predetermined duration (stored in the memory 16) such that the NOx regeneration process is performed only for the time required to remove the NOx groups stored in the NOx-Trap catalytic converter 11.

[0017] During the NOx regeneration process of the NOx-Trap catalytic converter 11, the control unit 15 monitors the signal from the multisensor 14; in particular, if no transition in the signal from the multisensor 14 from lean to rich is detected during the NOx regeneration process, then it is assumed that the actual duration of the NOx regeneration process matches the predetermined duration and that thus the storage capacity C of the NOx-Trap catalytic converter 11 is unchanged whereas, if a transition in the signal from the multisensor 14 from lean to rich is detected during the NOx regeneration process, then the actual duration of the NOx regeneration process was less than the predetermined duration and thus the NOx storage capacity C has clearly diminished. The signal from the multisensor 14 is significant because, while the reduction process of the NOx groups stored in the NOx-Trap catalytic converter 11 is under way, there is an excess of oxygen (relative to the stoichiometric value) in the exhaust gases downstream from the NOx-Trap catalytic converter 11, said excess arising from the reduction of the NOx groups, whereas once the reduction process of the NOx groups stored in the NOx-Trap catalytic converter 11 is complete, there is a deficit of oxygen (relative to the stoichiometric value) in the exhaust gases downstream from the NOx-Trap catalytic converter 11 because a rich mixture is supplied to the cylinders 2 during the regeneration process. It is clear from the above description that the signal from the multisensor 14 can also be used for estimating the actual duration of a NOx regeneration process, because, if no transition in the signal from the multisensor 14 from lean to rich is detected during the NOx regeneration process, then it is assumed that the actual value of the duration of the NOx regeneration process matches the predicted value whereas, if a transition in the signal from the multisensor 14 from lean to rich is detected during the NOx regeneration process, then the actual value of the duration of the NOx regeneration process is less than the calculated value and is equal to the period of time that has elapsed between the moment at which the regeneration process was initiated and the moment at which the transition in the signal from the multisensor 14 occurred.

[0018] If, during the NOx regeneration process of the NOx-Trap catalytic converter 11, the control unit 15 detects the above-described anomalous transition in the signal from the multisensor 14, then the control unit 15 attempts to determine the cause that led to the degradation of the trapping capacity C of the NOx-Trap catalytic converter 11 and thus attempts, if possible, to remedy such degradation. In particular, when the control unit 15 detects the anomalous transition in the signal from the multisensor 14, then the control unit 15 increases the operating temperature (indicatively by a step of 20-40° C.) of the NOx-Trap catalytic converter 11 by acting in known manner on the control of the throttle valve 7 and the injector 8 and awaits performance of the subsequent NOx regeneration process; if, during the subsequent NOx regeneration process, the control unit 15 again detects the anomalous transition in the signal from the multisensor 14, then the control unit 15 further increases the operating temperature (indicatively by a step of 20-40° C.) of the NOx-Trap catalytic converter 11 and awaits performance of the subsequent NOx regeneration process. Said process of increasing the operating temperature of the NOx-Trap catalytic converter 11 is continued in cycles not until the operating temperature of the NOx-Trap catalytic converter 11 reaches a predetermined limiting value, but instead until the control unit 15 ceases to detect the anomalous transition in the signal from the multisensor 14; in this latter case, i.e., if the increase in the operating temperature of the NOx-Trap catalytic converter 11 has led to the disappearance of the anomalous transition in the signal from the multisensor 14, then the control unit 15 increases the minimum value of the operating temperature of the NOx-Trap catalytic converter 11 stored in the memory 16 by a predetermined quantity because the reduction in the trapping capacity C of the NOx-Trap catalytic converter 11 is essentially due to thermal degradation and can be at least partly offset by increasing the operating temperature of the NOx-Trap catalytic converter 11. Obviously, the control unit 15 does not increase the minimum value of the operating temperature of the NOx-Trap catalytic converter 11 beyond a predetermined threshold value in order to maintain an acceptably wide operating temperature range for the NOx-Trap catalytic converter 11.

[0019] If, on the other hand, the increase in the operating temperature of the NOx-Trap catalytic converter 11 did not lead to the disappearance of the anomalous transition in the signal from the multisensor 14, then the control unit 15 performs an unscheduled desulfation process with a temperature value for the NOx-Trap catalytic converter 11 and an average value for ratio equal to the corresponding values used during the preceding desulfation processes. On completion of the unscheduled desulfation process, the control unit 15 awaits performance of the subsequent NOx regeneration process.

[0020] If, during the subsequent NOx regeneration process, the anomalous transition in the signal from the multisensor 14 no longer occurs, then the control unit 15 increments the temperature value of the NOx-Trap catalytic converter 11 and decrements the average value for ratio used during future desulfation processes because the reduction in the trapping capacity C of the NOx-Trap catalytic converter 11 is essentially due to the formation of particularly tenacious sulfur crystals. Obviously, the control unit 15 does not increase the temperature value for the NOx-Trap catalytic converter 11 nor does it decrement the average value for ratio used during desulfation processes beyond the respective predetermined thresholds.

[0021] If, on the other hand, the anomalous transition in the signal from the multisensor 14 still occurs in the subsequent NOx regeneration process, then the control unit 15 assumes that this phenomenon is due to irreversible degradation of the NOx-Trap catalytic converter 11 and thus reduces the estimated trapping capacity C of the NOx-Trap catalytic converter 11 used in the storage model for the NOx-Trap catalytic converter 11 by a predetermined amount.

[0022] According to another embodiment, the predicted duration of the NOx regeneration process of the NOx-Trap catalytic converter 11 is not assumed to be equal to a predetermined value stored in the memory 16, but is calculated before performing the NOx regeneration process by using the storage model for the NOx-Trap catalytic converter 11 and using the model of the production of NOx and SOx group by the engine 1 in such a manner that the NOx regeneration process only lasts for the time strictly necessary to remove the NOx groups trapped in the NOx-Trap catalytic converter 11.

[0023] From the above explanation, it is clear the control unit 15, before reducing the estimated trapping capacity C of the NOx-Trap catalytic converter 11 used in the storage model for the NOx-Trap catalytic converter 11, attempts to identify the cause of the degradation of the trapping capacity C of the NOx-Trap catalytic converter 11 and attempts, if possible, to remedy such degradation; by using this method, it is possible to avoid underestimating the actual trapping capacity C of the NOx-Trap catalytic converter 11 and thus to avoid increasing the frequency of NOx regeneration processes beyond what is strictly necessary.

Claims

1. Method for estimating the degradation of the trapping capacity of a NOx-Trap catalytic converter, which method provides for performing a first NOx regeneration process of a predetermined duration, determining whether the actual duration of the first NOx regeneration process is equal to the predetermined duration and assuming that the trapping capacity is unchanged if the actual duration of the first NOx regeneration process is equal to the predetermined duration; the method being characterised in that, if the actual duration of the first NOx regeneration process is less than the predetermined duration, at least one corrective action is performed in order to attempt to counteract the degeneration of the NOx-Trap catalytic converter, a subsequent NOx regeneration process is performed, it is determined whether the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then new characteristic operating parameters for the corrective action are used for the subsequent life of the NOx-Trap catalytic converter whereas, if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, the estimated trapping capacity (C) of the NOx-Trap catalytic converter is reduced.

2. Method according to claim 1, in which the signal of an ON/OFF type lambda sensor arranged upstream from the NOx-Trap catalytic converter is used to determine whether the actual duration of a NOx regeneration process is equal to the predetermined duration.

3. Method according to claim 2, in which, if no transition in the signal from the lambda sensor is detected during the NOx regeneration process, then it is assumed that the actual duration of the NOx regeneration process is equal to the predetermined duration whereas, if a transition in the signal from the lambda sensor is detected during the NOx regeneration process, then it is assumed that the actual duration of the NOx regeneration process is less than the predetermined duration.

4. Method according to claim 1, in which the corrective action provides for increasing the operating temperature of the NOx-Trap catalytic converter; if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then the minimum value of the operating temperature of said NOx-Trap catalytic converter is increased for the subsequent life of the NOx-Trap catalytic converter.

5. Method according to claim 4, in which the minimum value of the operating temperature of the NOx-Trap catalytic converter is not increased beyond a respective predetermined threshold value.

6. Method according to claim 4, in which the operating temperature of the NOx-Trap catalytic converter is increased by means of a number of successive increments of a determined size; after each increment, the performance of a subsequent NOx regeneration process is awaited and, if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, then a further increment is performed whereas, if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then incrementation of the operating temperature of the NOx-Trap catalytic converter is ceased and the minimum value of the operating temperature of said NOx-Trap catalytic converter is increased.

7. Method according to claim 6, in which the value of the operating temperature of the NOx-Trap catalytic converter is not increased beyond a respective predetermined threshold value.

8. Method according to claim 1, in which the corrective action provides for performing an unscheduled desulfation process and, on completion of the unscheduled desulfation process, awaiting performance of a subsequent NOx regeneration process; if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then the temperature value of the NOx-Trap catalytic converter is incremented and the average value for ratio used during future desulfation processes is decremented.

9. Method according to claim 8, in which the temperature value of the NOx-Trap catalytic converter and the average value for ratio used during the desulfation processes are not modified beyond respective predetermined threshold values.

10. Method according to claim 1, in which the corrective action provides for increasing the operating temperature of the NOx-Trap catalytic converter; if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then the minimum value of the operating temperature of said NOx-Trap catalytic converter is increased for the subsequent life of the NOx-Trap catalytic converter; if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, then an unscheduled desulfation process is performed and, on completion of the unscheduled desulfation process, performance of a subsequent NOx regeneration process is awaited; if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then the temperature value of the NOx-Trap catalytic converter is incremented and the average value for ratio used during future desulfation processes is decremented; if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, then the estimated trapping capacity of the NOx-Trap catalytic converter is reduced.

11. Method according to claim 10, in which the operating temperature of the NOx-Trap catalytic converter is increased by means of a number of successive increments of a determined size; after each increment, the performance of a subsequent NOx regeneration process is awaited and, if the actual duration of the subsequent NOx regeneration process is less than the predetermined duration, then a further increment is performed whereas, if the actual duration of the subsequent NOx regeneration process is equal to the predetermined duration, then incrementation of the operating temperature of the NOx-Trap catalytic converter is ceased and the minimum value of the operating temperature of the NOx-Trap catalytic converter is increased.

12. Method according to claim 11, in which the value of the operating temperature of the NOx-Trap catalytic converter is not increased beyond a respective predetermined threshold value.

13. Method according to claim 1, in which the predicted value for the duration of the NOx regeneration process is calculated using a storage model of the NOx-Trap catalytic converter, said model being based on an estimate of the trapping capacity of the NOx-Trap catalytic converter, such that the NOx regeneration process only lasts for the time that is strictly necessary to remove the NOx groups trapped in the NOx-Trap catalytic converter.