WO2010128564A1

WO2010128564A1 - Exhaust gas purifying device for internal combustion engine

Info

Publication number: WO2010128564A1
Application number: PCT/JP2009/059026
Authority: WO
Inventors: 美才治悠樹; 福田光一朗; 井上三樹男
Original assignee: トヨタ自動車株式会社
Priority date: 2009-05-08
Filing date: 2009-05-08
Publication date: 2010-11-11

Abstract

In an engine exhaust passage, there is arranged a NOx occluding/reducing catalyst for occluding NOx contained in inflow exhaust gases, when the inflow exhaust gases have a lean air/fuel ratio, and for releasing and reducing the occluded NOx when the air/fuel ratio of the inflow exhaust gases becomes rich. In order to release and reduce partially the NOx occluded in the NOx occluding/reducing catalyst, an enriching operation is performed to enrich the air/fuel ratio of the inflow exhaust gases temporarily. The NOx occluded quantity of the NOx occluding/reducing catalyst is calculated so that the enriching operation is controlled on the basis of the NOx occluded quantity. At the time of the enriching operation, the reduction of the NOx occluded quantity per unit time is calculated on the basis of the inflow reducer quantity that is the reducer quantity in the inflow exhaust gases, the NOx occluded quantity, and the catalyst temperature that is the temperature of the NOx occluding/reducing catalyst.

Description

Exhaust gas purification device for internal combustion engine

The present invention relates to an exhaust purification device for an internal combustion engine.

When the air-fuel ratio of the inflowing exhaust gas is lean, a NOx storage reduction catalyst that stores NOx contained in the exhaust gas and releases and reduces the stored NOx when the air-fuel ratio of the inflowing exhaust gas becomes rich is placed in the engine exhaust passage. Arranged internal combustion engines are known. In this internal combustion engine, NOx generated when combustion is performed under a lean air-fuel ratio is stored in the NOx storage reduction catalyst. On the other hand, for example, when the NOx occlusion amount of the NOx occlusion reduction catalyst increases, combustion is temporarily performed under a rich air-fuel ratio, and the air-fuel ratio of the inflowing exhaust gas is temporarily made rich, whereby the NOx occlusion reduction catalyst NOx is released and reduced.
In this case, if the air-fuel ratio of the inflowing exhaust gas is switched rich so that all NOx stored in the NOx storage reduction catalyst is released and reduced, a very large amount of fuel is consumed, and the fuel consumption rate increases. End up.
Therefore, an internal combustion engine is also known in which the air-fuel ratio of the inflowing exhaust gas is switched to be rich so that NOx occluded in the NOx occlusion reduction catalyst is partially released and reduced. In this way, it is possible to reduce the amount of fuel consumed for releasing and reducing NOx.
However, in this case, the NOx occlusion amount does not become zero even if the air-fuel ratio of the inflowing exhaust gas is switched to rich. Therefore, in order to accurately calculate the NOx occlusion amount, it is necessary to accurately calculate the decrease in the NOx occlusion amount when the air-fuel ratio of the inflowing exhaust gas is switched to rich.
Therefore, the air-fuel ratio of the inflowing exhaust gas can be calculated from a functional equation using the reducing agent concentration and oxygen concentration in the inflowing exhaust gas, the reducing agent utilization factor determined according to the catalyst temperature and the inflowing exhaust gas flow rate, and the amount of inflowing exhaust gas. An internal combustion engine that calculates a decrease in the NOx occlusion amount when rich is known (see Patent Document 1).
Further, an internal combustion engine is known that calculates a decrease in the NOx occlusion amount when the air-fuel ratio of the inflowing exhaust gas is rich based on the air-fuel ratio of the inflowing exhaust gas and the catalyst temperature (see Patent Document 2).

JP 2004-293338 A JP 2006-29331 A

However, in all cases, however, the actual decrease in the amount of NOx occlusion cannot always be calculated accurately. If a complicated calculation model or the like is used, the decrease in the NOx occlusion amount may be accurately calculated. However, it is not realistic to solve the complicated calculation model with a computer mounted on the vehicle.

According to the present invention, there is provided an exhaust purification device for an internal combustion engine having an exhaust passage, which is a NOx occlusion reduction catalyst disposed in the exhaust passage, and is contained in the inflow exhaust gas when the air-fuel ratio of the inflow exhaust gas is lean. NOx occlusion / reduction catalyst that occludes NOx and releases and reduces occluded NOx when the air-fuel ratio of the inflowing exhaust gas becomes rich, and NOx occluded in the NOx occlusion reduction catalyst partially releases and reduces A rich processing unit that performs rich processing for temporarily switching the air-fuel ratio of the inflowing exhaust gas to rich, a NOx storage amount calculation unit that calculates the NOx storage amount of the NOx storage reduction catalyst, and rich processing based on the NOx storage amount A rich processing control unit for controlling, an inflow reducing agent amount that is an amount of reducing agent in the inflowing exhaust gas, a NOx occlusion amount, and NO during the rich processing Storage reduction catalyst exhaust purifying apparatus for an internal combustion engine decrease per unit of time the NOx storage amount is calculated on the basis of the catalyst temperature is a temperature is provided.

The amount of NOx occlusion reduction when the air-fuel ratio of the inflowing exhaust gas is switched to rich can be calculated accurately and easily.

1 is an overall view of an internal combustion engine.
2 (A) and 2 (B) are cross-sectional views of the surface portion of the catalyst carrier of the NOx storage reduction catalyst.
3A and 3B are diagrams showing a map of the first correction coefficient KC1.
4A and 4B are diagrams showing a map of the second correction coefficient KC2.
5A and 5B are diagrams showing a map of the third correction coefficient KC3.
6A and 6B are time charts for explaining the relationship between the oxygen storage amount OXYst and the lean time tL.
FIG. 7 is a time chart for explaining how to supply fuel.
FIG. 8 is a diagram showing a map of index IDXeg.
FIG. 9 is a diagram showing a map of index IDXet.
10 is a diagram showing a map of index IDXsv.
FIG. 11 is a diagram showing a map of index IDXsp.
12 (A) and 12 (B) are diagrams showing a map of the NOx occlusion speed Vst.
FIG. 13 is a diagram showing a map of the NOx passage amount QNP.
FIG. 14 is a time chart for explaining the control of the rich process based on the rich process request index IDX.
FIGS. 15A and 15B are time charts for explaining control of rich processing based on the excess fuel ratio FER.
FIG. 16 is a diagram showing a map of a second fuel amount portion QFE2 during the purge process.
FIGS. 17A and 17B are time charts for explaining the success and failure of the purge process.
FIG. 18 is a flowchart showing an exhaust purification control routine.
FIG. 19 is a flowchart showing a rich process routine.
FIG. 20 is a flowchart showing a purge processing routine.
FIG. 21 is a flowchart showing a control routine for the rich flag XR.
FIG. 22 is a flowchart showing a control routine for a purge flag XP.
FIG. 23 is a flowchart showing a control routine for a purge flag XP.
FIG. 24 is a flowchart showing a routine for calculating the NOx occlusion amount NOXst.
FIG. 25 is a flowchart showing a routine for calculating the NOx occlusion amount NOXst.
FIG. 26 is a flowchart showing a routine for calculating a decrease NOXrd.
FIG. 27 is a flowchart showing a control routine for a purge failure flag XPF.
FIG. 28 is a flowchart showing a routine for calculating an excess fuel ratio FER.
FIG. 29 is a flowchart showing a routine for calculating a rich process request index IDX.
FIG. 30 is an overall view showing another embodiment of the compression ignition type internal combustion engine.
FIG. 31 is a time chart for explaining another embodiment of the operation for calculating the NOx occlusion amount NOXst.
FIG. 32 is a flowchart showing another embodiment of a routine for calculating the NOx occlusion amount NOXst.
FIG. 33 is a flowchart showing a routine for calculating the NOx occlusion amount NOXst.
FIG. 34 is a time chart for explaining a method of estimating a reduction amount RNrd by rich processing.

FIG. 1 shows a case where the present invention is applied to a compression ignition type internal combustion engine. However, the present invention can also be applied to a spark ignition internal combustion engine.
Referring to FIG. 1, 1 is an engine body, 2 is a combustion chamber of each cylinder, 3 is an electromagnetically controlled fuel injection valve for injecting fuel into each

combustion chamber

2, 4 is an intake manifold, and 5 is an exhaust manifold. Respectively. The intake manifold 4 is connected to the outlet of the compressor 7 c of the exhaust turbocharger 7 via the intake duct 6, and the inlet of the compressor 7 c is sequentially connected to the air flow meter 9 and the air cleaner 10 via the intake introduction pipe 8. An electrically controlled throttle valve 11 is disposed in the intake duct 6, and a cooling device 12 for cooling intake air flowing through the intake duct 6 is disposed around the intake duct 6. On the other hand, the exhaust manifold 5 is connected to the inlet of the exhaust turbine 7 t of the exhaust turbocharger 7, and the outlet of the exhaust turbine 7 t is connected to the exhaust aftertreatment device 20.
Each fuel injection valve 3 is connected to a common rail 14 through a fuel supply pipe 13, and this common rail 14 is connected to a fuel tank 16 through an electrically controlled fuel pump 15 having a variable discharge amount. The fuel in the fuel tank 16 is supplied into the common rail 14 by the fuel pump 15, and the fuel supplied into the common rail 14 is supplied to the fuel injection valve 3 through each fuel supply pipe 13. A fuel pressure sensor (not shown) for detecting the fuel pressure in the common rail 14 is attached to the common rail 14 so that the fuel pressure in the common rail 14 matches the target pressure based on a signal from the fuel pressure sensor. The fuel discharge amount of the fuel pump 15 is controlled.
The exhaust manifold 5 and the intake manifold 4 are connected to each other via an exhaust gas recirculation (hereinafter referred to as EGR) passage 17, and an electrically controlled EGR control valve 18 is disposed in the EGR passage 17. A cooling device 19 for cooling the EGR gas flowing in the EGR passage 17 is disposed around the EGR passage 17.
The exhaust aftertreatment device 20 includes an exhaust pipe 21 connected to an outlet of the exhaust turbine 7t. The exhaust pipe 21 is connected to a casing 22, and the casing 22 is connected to an exhaust pipe 23. A relatively small capacity three-way catalyst 24 is accommodated on the upstream side in the casing 22, and a relatively large capacity NOx storage reduction catalyst 25 is accommodated on the downstream side. The three-way catalyst 24 and the NOx storage reduction catalyst 25 may be housed in separate casings. Further, the three-way catalyst 24 may be omitted.
The exhaust pipe 21 is provided with a temperature sensor 26 that detects the temperature of the exhaust gas flowing into the NOx storage reduction catalyst 25 and an air-fuel ratio sensor 27 that detects the air-fuel ratio of the exhaust gas flowing into the casing 22. In addition, a temperature sensor 28 that detects a catalyst temperature that is the temperature of the NOx storage reduction catalyst 25 and a differential pressure sensor 29 that detects a differential pressure across the casing 22 are attached to the casing 22. Further, an air-fuel ratio sensor 30 for detecting the air-fuel ratio of exhaust gas flowing out from the casing 22 is attached to the exhaust pipe 23.
The electronic control unit 40 is composed of a digital computer, and is connected to each other by a bidirectional bus 41. A ROM (read only memory) 42, a RAM (random access memory) 43, a CPU (microprocessor) 44, an input port 45 and an output port 46 are connected. It comprises. The output voltages of the air flow meter 9, the

temperature sensors

26 and 28, the air-

fuel ratio sensors

27 and 30, and the differential pressure sensor 29 are input to the input port 45 via the corresponding AD converters 47. A load sensor 50 that generates an output voltage proportional to the amount of depression of the accelerator pedal 49 is connected to the accelerator pedal 49, and the output voltage of the load sensor 50 is input to the input port 45 via the corresponding AD converter 47. The Further, a crank angle sensor 51 that generates an output pulse every time the crankshaft rotates, for example, 30 degrees is connected to the input port 45. The CPU 44 calculates the engine speed based on the output pulse from the crank angle sensor 51. On the other hand, the output port 46 is connected to the fuel injection valve 3, the drive device for the throttle valve 11, the fuel pump 15, and the EGR control valve 18 via a corresponding drive circuit 48.
The NOx storage reduction catalyst 25 has a honeycomb structure and includes a plurality of exhaust gas flow passages separated from each other by thin partition walls. A catalyst carrier made of alumina, for example, is supported on both side surfaces of each partition wall. FIGS. 2A and 2B schematically show a cross section of the surface portion of the catalyst carrier 55. As shown in FIGS. 2 (A) and 2 (B), a noble metal catalyst 56 is dispersedly supported on the surface of the catalyst carrier 55, and a layer of NOx absorbent 57 is further provided on the surface of the catalyst carrier 55. Is formed.
In the embodiment according to the present invention, at least one selected from platinum Pt, palladium Pd, osmium Os, gold Au, rhodium Rh, iridium Ir, and ruthenium Ru is used as the noble metal catalyst 56, and the NOx absorbent 57 is configured as a component. For example, at least one selected from alkali metals such as potassium K, sodium Na and cesium Cs, alkaline earth such as barium Ba and calcium Ca, and rare earths such as lanthanum La and yttrium Y is used.
When the ratio of the reducing agent such as air and fuel supplied into the intake passage, the combustion chamber 2 and the exhaust passage upstream of a certain position in the exhaust passage is referred to as the air-fuel ratio of the exhaust gas at that position, NOx The absorbent 57 absorbs NOx when the air-fuel ratio of the inflowing exhaust gas is lean, and performs the NOx absorption / release action of releasing the absorbed NOx when the oxygen concentration in the inflowing exhaust gas decreases.
That is, the case where platinum Pt is used as the noble metal catalyst 56 and barium Ba is used as a component constituting the NOx absorbent 57 will be described as an example. When the air-fuel ratio of the inflowing exhaust gas is lean, that is, the oxygen concentration in the inflowing exhaust gas When NO is high, NO contained in the exhaust gas is oxidized on the platinum Pt 56 as shown in FIG.₂And then absorbed into the NOx absorbent 57 and barium carbonate BaCO.₃Nitrate ion NO while binding₃ ⁻In the form of NOx absorbent 57. In this way, NOx is absorbed in the NOx absorbent 57. As long as the oxygen concentration in the inflowing exhaust gas is high, NO on the surface of platinum Pt56₂NOx is not generated unless the NOx absorption capacity of the NOx absorbent 57 is saturated.₂Is absorbed in the NOx absorbent 57 and nitrate ions NO.₃ ⁻Is generated.
In contrast, when the air-fuel ratio of the inflowing exhaust gas is made rich, the oxygen concentration in the inflowing exhaust gas decreases, so the reaction is reversed (NO₃ ⁻→ NO₂Thus, as shown in FIG. 2B, nitrate ion NO in the NOx absorbent 57₃ ⁻Is NO₂In the form of NOx absorbent 57. Next, the released NOx is reduced by a reducing agent such as HC or CO contained in the exhaust gas.
On the other hand, the three-way catalyst 24 also has a honeycomb structure and includes a plurality of exhaust gas flow passages separated from each other by thin partition walls. A catalyst carrier made of, for example, alumina is supported on both surfaces of each partition wall, and a noble metal such as platinum Pt, palladium Pd, and rhodium Rh and an oxygen storage material such as cerium oxide are supported on the surface of the catalyst carrier. It is supported.
This three-way catalyst 24 simultaneously purifies HC, CO, NOx contained in the exhaust gas when the air-fuel ratio of the inflowing exhaust gas is substantially the stoichiometric air-fuel ratio.
The three-way catalyst 24 constitutes an oxygen storage catalyst that stores oxygen contained in the exhaust gas when the air-fuel ratio of the inflowing exhaust gas is lean and releases the stored oxygen when the air-fuel ratio of the inflowing exhaust gas becomes rich. . That is, the case where the oxygen storage material is made of cerium Ce will be described as an example. When the air-fuel ratio of the inflowing exhaust gas is lean, the oxygen molecules O contained in the exhaust gas are described.₂Is CeO₂(Ce₂O₃-> 2CeO₂). In contrast, when the air-fuel ratio of the inflowing exhaust gas becomes rich, the reaction proceeds in the reverse direction (2CeO₂-> Ce₂O₃), Oxygen molecule O₂Is released.
Now, in the embodiment according to the present invention, combustion is usually performed under a lean air-fuel ratio. In this case, since the air-fuel ratio of the inflowing exhaust gas to the NOx storage reduction catalyst 25 is lean, NOx in the inflowing exhaust gas is stored in the NOx storage reduction catalyst 25 at this time. However, if the engine operation is continued, the NOx occlusion amount of the NOx occlusion reduction catalyst 25 increases, and eventually the NOx occlusion reduction catalyst 25 cannot occlude NOx.
Therefore, in the embodiment according to the present invention, the air-fuel ratio of the inflowing exhaust gas is temporarily enriched before the NOx occlusion reduction catalyst 25 is saturated with NOx, thereby releasing NOx from the NOx occlusion reduction catalyst 25.₂I'm trying to reduce it. As a result, the NOx occlusion amount is reduced.
Specifically, in order to partially release and reduce NOx occluded in the NOx occlusion reduction catalyst 25, a rich process for temporarily changing the air-fuel ratio of the inflowing exhaust gas to rich is performed. In this case, in order to switch the air-fuel ratio of the inflowing exhaust gas to rich, combustion is temporarily performed under the rich air-fuel ratio while decreasing the intake air amount and increasing the EGR gas amount. As a result, a large amount of carbon monoxide CO is contained in the exhaust gas, and this CO mainly acts as a reducing agent.
A lot of fuel is required to perform rich processing. Therefore, in order to suppress fuel consumption, it is appropriately determined whether or not rich processing should be started during engine operation under a lean air-fuel ratio, and whether or not rich processing should be continued during rich processing. It is necessary to judge appropriately.
Therefore, in the embodiment according to the present invention, the NOx occlusion amount NOXst of the NOx occlusion reduction catalyst 25 is calculated, and the rich process is controlled based on the NOx occlusion amount NOXst.
In the embodiment according to the present invention, the NOx occlusion amount NOXst is repeatedly calculated based on the following formula E1.

Here, NOXin and NOXrd respectively represent an increment and a decrease of the NOX occlusion amount NOXst per unit time.
The increment NOXin is set to the inflow NOx amount that is the NOx amount in the inflow exhaust gas when the air-fuel ratio of the inflow exhaust gas is lean, and is set to zero when the air-fuel ratio of the inflow exhaust gas is rich or the stoichiometric air-fuel ratio. Here, the inflow NOx amount is calculated based on the engine operating state, for example, the engine load and the engine speed. Alternatively, a NOx sensor for detecting the inflow NOx amount is disposed in the exhaust passage upstream of the NOx storage reduction catalyst 25, and an increase NOXin when the air-fuel ratio of the inflowing exhaust gas is lean is obtained based on the output of the NOx sensor. It may be.
In contrast, the decrease NOXrd is zero when the air-fuel ratio of the inflowing exhaust gas is lean, and is calculated based on the following equation E2 when the air-fuel ratio of the inflowing exhaust gas is rich or the stoichiometric air-fuel ratio.

Here, k0 is a constant, AFS is the stoichiometric air-fuel ratio, AFI is the air-fuel ratio of the inflowing exhaust gas detected by the air-fuel ratio sensor 27, E is the activation energy, R is the gas constant, and T is detected by the temperature sensor 28. Catalyst temperature, Δt represents a calculation time interval, and KC represents a correction coefficient.
Mathematical formula E2 is derived as follows. That is, when CO is used as the reducing agent, the NOx reduction reaction is represented by the following reaction formula R1.

Therefore, the time change dNOXst / dt of the NOx occlusion amount when the air-fuel ratio of the inflowing exhaust gas is rich is expressed by the following equation E3 in consideration of the reaction rate of the reaction equation R1.

Here, k represents a constant, and QCO represents an inflow CO amount that is the CO amount in the inflow exhaust gas.
When the equation E3 is discretized using i representing the number of calculations, the following equation E4 is obtained.

Here, the inventors of the present application have confirmed that the inflow CO amount QCO is represented by the air-fuel ratio AFI of the inflow exhaust gas. Specifically, the inflow CO amount QCO increases as the richness of the air-fuel ratio AFI of the inflowing exhaust gas, that is, the deviation (AFS-AFI) of the air-fuel ratio AFI of the inflowing exhaust gas with respect to the stoichiometric air-fuel ratio AFS increases.
Therefore, since the left side of the equation E4 is equal to the negative value (−NOXrd) of the decrease NOXrd, the equation E2 is derived by introducing the correction coefficient KC into the equation E4.
Using the above-described equation E2, it is possible to easily and accurately calculate the decrease NOXrd during the rich process. In fact, it has been experimentally confirmed that the decrease NOXrd can be calculated with an error of about 3 percent.
During the rich process, the decrease NOXrd is repeatedly calculated using the above-described equation E2, and the NOx occlusion amount NOXst is repeatedly updated by the decrease NOXrd.
Thus, in the embodiment according to the present invention, the decrease NOXrd is calculated based on the inflow CO amount QCO, the NOx occlusion amount NOXst, and the catalyst temperature T.
One thing to be noted here is that the decrease NOXrd is calculated based on the NOx occlusion amount NOXst. As described at the beginning, the conventional calculation method cannot always calculate the decrease NOXrd accurately. Therefore, as long as it is based on such an inaccurate NOx occlusion amount NOXst, the decrease NOXrd cannot always be calculated accurately. That is, it has not been possible to calculate the decrease NOXrd based on the NOx occlusion amount NOXst.
Another notable feature is that the NOx occlusion amount NOXst, that is, the decrease NOxrd during the rich process is calculated without obtaining the outflow NOx amount that is the amount of NOx flowing out from the NOx occlusion reduction catalyst 25. When the air-fuel ratio of the inflowing exhaust gas is rich, it becomes difficult to accurately detect the outflow NOx amount with the NOx sensor disposed in the exhaust passage of the NOx storage reduction catalyst 25. In the embodiment according to the present invention, since the amount of outflow NOx is not required, the decrease NOXrd can be accurately calculated. Further, it is not necessary to provide a NOx sensor downstream of the NOx storage reduction catalyst 25.
The above formula E3 can be applied to a reducing agent in which the molar ratio of NOx to the reducing agent in the reduction reaction is theoretically 1: 2. Hydrogen H as such a reducing agent₂Is mentioned. In this case, the reduction reaction of NOx is represented by the following reaction formula R2.

Therefore, when the amount of hydrogen in the inflowing exhaust gas is represented by QH, the decrease NOXrd in this case is calculated from the following equation E5.

Therefore, in general terms, the decrease NOXrd is calculated based on the following terms using the inflow reductant amount QRED, the NOx occlusion amount NOXst, and the catalyst temperature T, which is the amount of reductant in the inflow exhaust gas. .

Furthermore, the concept of calculating the decrease NOXrd based on the inflow reducing agent amount QRED, the NOx occlusion amount NOXst, and the catalyst temperature T is that the reducing agent is CO, H₂It can be applied not only to other materials such as hydrocarbon HC.
Therefore, when further generalized, the reduced amount NOXrd is calculated based on the inflow reducing agent amount QRED, the NOx occlusion amount NOXst, and the catalyst temperature T.
Next, the correction coefficient KC in the equation E2 will be described.
The correction coefficient KC is calculated based on the following formula E6.

Here, KC1, KC2, and KC3 represent a first correction coefficient, a second correction coefficient, and a third correction coefficient, respectively. Each correction coefficient KC1, KC2, KC3 is set to 1.0 when there is no need to correct each.
The first correction coefficient KC1 is used to correct the amount of decrease NOXrd based on the CO passage amount that is the amount of CO passing through the NOx storage reduction catalyst 25 without reducing NOx during the rich process. That is, as the CO passing amount increases, the decrease NOXrd decreases. Therefore, the decrease NOXrd is corrected to decrease by the first correction coefficient KC1. As shown in FIG. 3A, the first correction coefficient KC1 becomes smaller as the intake air amount Ga detected by the air flow meter 9 increases, and the difference between the front and rear of the casing 22 detected by the differential pressure sensor 29 is increased. The pressure dP decreases as the pressure dP decreases. That is, the CO passage amount increases as the inflowing exhaust gas amount increases. Further, as the degree of clogging of the casing 22 in which the three-way catalyst 24 and the NOx storage reduction catalyst 25 are housed becomes small, the residence time of the exhaust gas in the casing 22 is shortened, and thus the amount of CO passing is increased. Here, the inflow exhaust gas amount is represented by the intake air amount Ga. Further, the degree of clogging of the casing 22 is represented by the differential pressure dP. Therefore, the intake air amount Ga and the differential pressure dP are detected, and the first correction coefficient KC1 is calculated based on the intake air amount Ga and the differential pressure dP. Note that the first correction coefficient KC1 is stored in the ROM 42 in the form of a map shown in FIG. 3B as a function of the intake air amount Ga and the differential pressure dP.
The second correction coefficient KC2 is used to correct the amount of decrease NOXrd based on the degree of deterioration of the NOx storage reduction catalyst 25. That is, the decrease NOXrd decreases as the degree of deterioration increases. Therefore, the decrease NOXrd is corrected to decrease by the second correction coefficient KC2. As shown in FIG. 4A, the second correction coefficient KC2 decreases as the integrated value ST of the catalyst temperature increases, and decreases as the SOx storage amount SOXst of the NOx storage reduction catalyst 25 increases. That is, as the integrated value ST of the catalyst temperature increases, the degree of thermal deterioration of the NOx storage reduction catalyst 25 increases. Further, as the SOx occlusion amount SOXst of the NOx occlusion reduction catalyst 25 increases, the NOx occlusion capacity of the NOx occlusion reduction catalyst 25 decreases, and therefore the degree of deterioration increases. Therefore, the integrated value ST of the catalyst temperature and the SOx occlusion amount SOXst are calculated, and the second correction coefficient KC2 is calculated based on the integrated value ST of the catalyst temperature and the SOx occlusion amount SOXst. The second correction coefficient KC2 is stored in the ROM 42 in the form of a map shown in FIG. 4B as a function of the integrated value ST of the catalyst temperature and the SOx occlusion amount SOXst.
The third correction coefficient KC3 corrects the inflow CO amount QCO to be reduced based on the CO consumption in the exhaust gas consumed by the three-way catalyst 24 during the rich process, and consequently reduces the decrease NOXrd to a decrease. Is to do. That is, CO in the exhaust gas is oxidized or consumed by oxygen released from the three-way catalyst 24 during the rich process. As this CO consumption increases, the inflow CO amount decreases and the decrease NOXrd decreases. Therefore, the decrease NOXrd is corrected to decrease by the third correction coefficient KC3. As shown in FIG. 5A, the third correction coefficient KC3 decreases as the intake air amount Ga decreases, and the air-fuel ratio AFI of the inflowing exhaust gas is maintained lean until the rich process is started. As the lean time tL, which is the waiting time, becomes longer, it becomes smaller. That is, the CO consumption increases as the inflowing exhaust gas amount represented by the intake air amount Ga decreases. Further, the CO consumption increases as the oxygen storage amount OXYst of the three-way catalyst 24 increases when the rich process is started. Here, the oxygen storage amount OXYst of the three-way catalyst 24 when the rich process is started is represented by the lean time tL. The third correction coefficient KC3 is stored in the ROM 42 in the form of a map shown in FIG. 5B as a function of the intake air amount Ga and the lean time tL.
The lean time tL will be further described. When the lean time tL is long as shown in FIG. 6 (A), the oxygen storage amount OXYst of the three-way catalyst 24 is increased, and then the rich process is shown as X. Is started. In contrast, when the lean time tL is short as shown in FIG. 6B, the rich process is started as indicated by X while the oxygen storage amount OXYst is small. Therefore, the lean time tL represents the oxygen storage amount OXYst when the rich process is started.
Therefore, the intake air amount Ga and the lean time tL are detected, and the third correction coefficient KC3 is calculated based on the intake air amount Ga and the lean time tL. The third correction coefficient KC3 is stored in the ROM 42 in the form of a map shown in FIG. 5B as a function of the intake air amount Ga and the lean time tL.
It should be noted that the oxygen storage reaction of the three-way catalyst 24 when the air-fuel ratio of the inflowing exhaust gas is lean is represented by the following reaction formula R3.

Therefore, the oxygen storage amount OXYst is calculated from the time change dOXYst / dt of the oxygen storage amount of the three-way catalyst 24 obtained in consideration of the reaction rate of the reaction formula R3, and the inflowing CO amount based on the calculated oxygen storage amount OXYst. May be corrected.
On the other hand, the constant k0 in the equation E2 is obtained in advance by experiments. That is, for example, the decrease NOXtd is measured under various NOx occlusion amounts NOXst and the catalyst temperature T. After that, the measured decrease NOXrd is substituted into Equation E2 to calculate k0, and by averaging these k0, a constant k0 is calculated. During the rich treatment, not only NOx reduction, but also NOx occlusion, NOx reduction by reducing agents other than CO in the exhaust gas, for example, hydrogen and hydrocarbons, NOx occlusion amount NOXst and catalyst temperature T, three-dimensional distribution, rich Changes in the catalyst temperature T during processing, NOx outflow from the NOx storage reduction catalyst 25 when the air-fuel ratio of the inflowing exhaust gas switches from lean to rich, NOx storage reduction when the air-fuel ratio of the inflowing exhaust gas is lean An outflow of NOx from the catalyst 25 occurs. Therefore, when the constant k0 is calculated as described above, various phenomena that may affect the decrease NOXrd can be comprehensively expressed by the equation E2. This means that the decrease NOXrd can be calculated easily and accurately with the formula E2.
Even if the air-fuel ratio AFI of the exhaust gas flowing into the casing 22 is rich, the exhaust gas flowing into the NOx storage reduction catalyst 25 may not become rich because of the oxygen storage capacity of the three-way catalyst 24. On the other hand, while the air-fuel ratio of the exhaust gas flowing into the NOx storage reduction catalyst 25 is rich and the NOx release and reduction action is being performed, the air-fuel ratio of the exhaust gas flowing out from the NOx storage reduction catalyst 25 is the stoichiometric air-fuel ratio. Or it has been confirmed that it is kept rich.
Therefore, in the embodiment according to the present invention, when the air-fuel ratio AFO of the outflow exhaust gas detected by the air-fuel ratio sensor 30 is lean, the increase NOXin is set to the inflow NOx amount, and the decrease NOXrd is made zero. Further, when the air-fuel ratio AFO of the outflow exhaust gas is the stoichiometric air-fuel ratio or rich, the increment NOXin is set to zero, and the decrease NOXrd is calculated using Equation E2.
Thus, in the embodiment according to the present invention, the NOx decrease amount NOXrd can be accurately calculated, and therefore the NOx occlusion amount NOXst can be accurately calculated.
However, a slight calculation error may be included in the decrease NOXrd or the NOx occlusion amount NOXst. For this reason, when the engine operation time becomes longer, the calculated NOXst may deviate from the actual NOx storage amount.
Therefore, in the embodiment according to the present invention, in order to release and reduce almost all of the NOx stored in the NOx storage reduction catalyst 25, a purge process for temporarily switching the air-fuel ratio of the inflowing exhaust gas to rich is performed. . If the NOx occlusion amount NOXst is returned to zero when almost all NOx in the NOx occlusion reduction catalyst 25 is released and reduced by the purge process, the NOx occlusion amount NOXst accurately represents the actual NOx occlusion amount.
In the purge process of the embodiment according to the present invention, as shown in FIG. 7 (i), in addition to the main fuel M for obtaining the engine output supplied around the compression top dead center (TDC), the end stage of the expansion stroke. Alternatively, additional fuel A is injected from the fuel injection valve 3 in the exhaust stroke, whereby the air-fuel ratio of the inflowing exhaust gas is switched to rich. In this case, the inflowing exhaust gas contains a large amount of HC, and HC mainly acts as a reducing agent. In the example shown in FIG. 7I, the additional fuel A hardly contributes to the engine output.
In this case, additional fuel A is supplied every time the cylinder enters the expansion stroke or the exhaust stroke. Therefore, during the purge process, the additional fuel A or the reducing agent is supplied into the combustion chamber 2 in a secondary and pulse form.
Alternatively, in order to switch the air-fuel ratio of the inflowing exhaust gas to be rich, additional fuel A may be injected from the fuel injection valve 3 immediately after the main fuel M is injected, as shown in (ii) of FIG. . In the example shown in FIG. 7 (ii), the additional fuel A partially burns in the combustion chamber 2, and thus partially contributes to the engine output. For this reason, a mechanism for absorbing an additional output from the additional fuel A, for example, an electric motor of a hybrid vehicle is required.
Alternatively, as shown in (iii) of FIG. 7, the main fuel M may be used to temporarily perform combustion under a rich air-fuel ratio. FIG. 7 (iv) shows a case where combustion is performed under a lean air-fuel ratio and no additional fuel is injected.
Even during the rich processing, additional fuel A can be injected as shown in FIG. 7 (i) or (ii) in order to switch the air-fuel ratio of the inflowing exhaust gas to rich.
Next, control of rich processing according to an embodiment of the present invention will be described.
In the embodiment according to the present invention, the rich processing request index IDX is repeatedly calculated when the air-fuel ratio of the inflowing exhaust gas is lean. The rich process request index IDX indicates the degree of necessity of the rich process, and indicates that the necessity of the rich process increases as the rich process request index IDX increases (0 ≦ IDX ≦ 1). In addition, the rich process is not performed when the rich process request index IDX is smaller than the threshold value TH, and the rich process is performed when the rich process request index IDX is larger than the threshold value TH.
The rich process request index IDX is calculated based on the following formula E7.

Here, IDX1 and IDX2 represent a first index and a second index, respectively.
When the decrease NOXrd per unit reducing agent (CO) amount is referred to as reducing agent utilization efficiency, the first index IDX1 is determined according to the reducing agent utilization efficiency when it is assumed that the rich process is performed. It is calculated based on the following formula E8 using IDXeg and IDXet.

The index IDXeg is obtained based on the inflow exhaust gas amount. That is, as the inflowing exhaust gas amount increases, the CO passage amount increases, and therefore the reducing agent utilization efficiency decreases. The index IDXeg is obtained in advance by standardizing the reducing agent utilization efficiency that varies according to the intake air amount Ga representing the inflowing exhaust gas amount (0 ≦ IDXeg ≦ 1). That is, as shown in FIG. 8, the index IDXeg increases as the intake air amount Ga decreases, and is maintained at 1 when the intake air amount Ga further decreases. The index IDXeg is stored in the ROM 42 in the form of a map shown in FIG. 8 as a function of the intake air amount Ga.
The index IDXet is obtained based on the catalyst temperature T. That is, the ease of the NOx reduction reaction in the NOx occlusion reduction catalyst 25 varies according to the catalyst temperature T, and therefore the reducing agent utilization efficiency varies. The index IDXet is obtained in advance by normalizing the reducing agent utilization efficiency that varies according to the catalyst temperature T (0 ≦ IDXet ≦ 1). That is, as shown in FIG. 9, the index IDXet increases as the catalyst temperature T increases when the catalyst temperature T is low, and decreases as the catalyst temperature T increases when the catalyst temperature T is high. The index IDXet is stored in the ROM 42 as a function of the catalyst temperature T in the form of a map shown in FIG.
On the other hand, the second index IDX2 is determined according to the NOx occlusion capacity of the NOx occlusion reduction catalyst 25, and is calculated based on the following formula E9 using the indices IDXsv and IDXsp.

When the ratio of the amount of NOx occluded in the NOx occlusion reduction catalyst 25 to the amount of NOx in the inflowing exhaust gas is referred to as NOx occlusion speed Vst (0 ≦ Vst ≦ 1), the index IDXsv is the NOx occlusion speed Vst. It is required based on. That is, NOx is less likely to be stored in the NOx storage reduction catalyst 25 as the NOx storage speed Vst decreases. As shown in FIG. 10, the index IDXsv increases from zero as the NOx occlusion speed Vst decreases from 1, and is maintained at 1 when it falls below the allowable lower limit LV (0 ≦ IDXsv ≦ 1). The index IDXsv is stored in the ROM 42 in the form of a map shown in FIG. 10 as a function of the NOx occlusion speed Vst.
The index IDXsp is obtained based on the NOx passage amount QNP that is the amount of NOx that passes through the NOx storage reduction catalyst when the air-fuel ratio of the inflowing exhaust gas is lean. That is, as the NOx passage amount QNP increases, NOx is less likely to be stored in the NOx storage reduction catalyst 25. As shown in FIG. 11, the index IDXsp increases as the NOx passage amount QNP increases, and is maintained at 1 when the allowable upper limit UN is exceeded (0 ≦ IDXsp ≦ 1). The index IDXsp is stored in the ROM 42 in the form of a map shown in FIG. 11 as a function of the NOx passage amount QNP.
Here, as shown in FIG. 12A, the NOx occlusion speed Vst increases as the NOx occlusion amount NOXst decreases. When the catalyst temperature T is low, the NOx occlusion speed Vst increases as the catalyst temperature T increases. When it is high, it decreases as the catalyst temperature T increases. The NOx occlusion speed Vst is stored in the ROM 42 in the form of a map shown in FIG. 12B as a function of the NOx occlusion amount NOXst and the catalyst temperature T. Note that the NOx occlusion speed Vst does not have to standardize the above-described ratio of the NOx amount.
On the other hand, the NOx passage amount QNP is stored in the ROM 42 in the form of a map shown in FIG. 13 as a function of the inflow NOx amount QNI, the NOx occlusion speed Vst, and the NOx passage rate RP. The NOx passage rate RP is the ratio of the NOx amount that has passed through the NOx storage reduction catalyst 25 to the inflow NOx amount when the air-fuel ratio of the inflowing exhaust gas is lean. For example, the ROM 42 as a function of the intake air amount Ga that represents the exhaust gas flow rate. Is stored within.
Note that the rich process request index IDX may be obtained based on one of the first index IDX1 and the second index IDX2. Further, the first index IDX1 may be obtained based on one of the index IDXeg and the index IDXet, or the second index IDX2 may be obtained based on either the index IDXsv or the index IDXsp. Also good.
When the index IDXet, IDXet is increased, the first index IDX1 is increased. When the index IDXsv, IDXsp is increased, the second index IDX2 is increased, so that the rich process request index IDX is increased. As described above, the rich process is performed when the rich process request index IDX becomes greater than the threshold value TH.
In this way, for example, when the reducing agent cannot be effectively used even if the NOx occlusion amount NOXst is large, the rich process is not started. Therefore, it is possible to reliably reduce NOx while suppressing consumption of the reducing agent or fuel.
On the other hand, the NOx occlusion speed Vst described above is repeatedly calculated during the rich process. Since the NOx occlusion amount NOXst gradually decreases during the rich process, the NOx occlusion speed Vst gradually increases. Next, the rich process is completed when the NOx occlusion speed Vst reaches a predetermined target NOx occlusion speed TVst.
The target NOx occlusion speed TVst is set to be smaller than 1, and the NOx occlusion amount NOXst is not reduced to zero by the rich process. As a result, NOx occluded in the NOx occlusion reduction catalyst 25 is partially released and reduced, so that consumption of fuel or reducing agent can be suppressed.
In summary, as shown by X in FIG. 14, when the rich process request index IDX becomes larger than the threshold value TH, the rich process is started, and when the NOx occlusion speed Vst reaches the target NOx occlusion speed TVst during the rich process, the rich process is started. The process will be completed.
In the embodiment according to the present invention, the rich process is controlled based on the excess fuel ratio FER.
This fuel excess ratio FER is repeatedly calculated based on the following formula E10.

Here, SQFO represents an operating fuel amount integrated value, and SQFE represents an excess fuel amount integrated value.
The operating fuel amount integrated value SQFO is an integrated value of the operating fuel amount QFO, which is the amount of fuel supplied for engine operation under a lean air-fuel ratio.
On the other hand, the excess fuel amount integrated value SQFE is an excess fuel amount QFE that is the amount of fuel supplied in addition to the operating fuel amount QFO in order to make the air-fuel ratio of the inflowing exhaust gas rich, that is, for rich processing and purge processing. Is an integrated value of. In the example shown in (i) or (ii) of FIG. 7, the operating fuel amount QFO corresponds to the amount of the main fuel M, and the excess fuel amount QFE corresponds to the amount of the additional fuel A.
Here, the excess fuel amount QFE includes the first fuel amount portion QFE1 for reducing the air-fuel ratio AFI of the inflowing exhaust gas from the lean air-fuel ratio obtained by the operating fuel amount QFO to the stoichiometric air-fuel ratio AFS, and the airflow of the inflowing exhaust gas. It is expressed as the sum of the second fuel amount portion QFE2 for making the fuel ratio AFI rich beyond the stoichiometric air-fuel ratio (QFE = QFE1 + QFE2). The second fuel amount portion QFE2 represents the amount of fuel that can substantially act as a reducing agent. In the rich process according to the embodiment of the present invention, the second fuel amount portion QFE2 is set so that the air-fuel ratio AFI of the inflowing exhaust gas becomes the target rich air-fuel ratio.
Fuel excess ratio FER represents the degree of deterioration of the fuel consumption rate due to rich processing or purge processing. That is, when the rich process or the purge process is started, the excess fuel amount integrated value SQFE gradually increases, so that the excess fuel ratio FER gradually increases. On the other hand, if the rich process and the purge process are not performed, the excess fuel amount integrated value SQFE does not increase, so the excess fuel ratio FER gradually decreases.
It is not preferable that the excess fuel ratio FER is excessively large. Therefore, in the embodiment according to the present invention, the rich process is interrupted when the excess fuel ratio FER exceeds the predetermined upper limit value UR1 during the rich process. As a result, excessive fuel consumption is prevented.
On the other hand, the excessive fuel ratio FER means that the rich process and the purge process have not been performed for a long time, which is also not preferable. Therefore, in the embodiment according to the present invention, when the excess fuel ratio FER becomes smaller than the predetermined lower limit value LR1, the rich process is performed regardless of the rich process request index IDX. As a result, it is possible to prevent NOx release and reduction action from being performed for a long time.
In summary, as shown by X in FIG. 15A, when the excess fuel ratio FER becomes smaller than the lower limit value LR1, rich processing is started regardless of the rich processing request index IDX, and Y in FIG. As shown, when the excess fuel ratio FER exceeds the upper limit value UR1 during the rich process, the rich process is interrupted regardless of the NOx release speed Vst.
Next, control of purge processing according to an embodiment of the present invention will be described.
If a purge process is performed when the inflowing exhaust gas amount is large, a large amount of fuel or reducing agent is required. Further, even if the purge process is performed when the catalyst temperature T is low, NOx cannot be sufficiently reduced. Further, even if the purge process is performed when the NOx occlusion amount NOXst is small, the fuel or the reducing agent cannot be effectively used for NOx reduction.
Therefore, in the embodiment according to the present invention, the intake air amount Ga representing the inflow exhaust gas amount is larger than the upper limit amount UG2, the catalyst temperature T is lower than the lower limit temperature T2, or the NOx occlusion amount NOXst is lower than the lower limit amount LN2. When the number is small, it is determined that the purge condition is not satisfied, and the purge process is prohibited.
In contrast, if the intake air amount Ga is smaller than the upper limit amount UG2, the catalyst temperature T is higher than the lower limit temperature LT2, and the NOx occlusion amount NOXst is larger than the lower limit amount LN2, it is determined that the purge condition is satisfied. The purge process is started.
When the purge process is started, the main fuel M and the additional fuel A are supplied to each cylinder as described above with reference to FIG. Here, the amount of the main fuel M and the amount of the additional fuel A correspond to the operating fuel amount QFO and the excess fuel amount QFE, respectively. The excess fuel amount QFE is the first fuel amount portion QFE1 and the second fuel amount. It is the sum of the partial QFE2.
In the embodiment according to the present invention, when the purge process is started, the second fuel amount partial integrated value SQFE2 is repeatedly calculated, and when the second fuel amount partial integrated value SQFE2 reaches the purge fuel amount QFP, the supply of additional fuel A is performed. Is stopped, and thus the purge process is completed. In this case, it can be considered that the purge fuel amount QFP is divided and supplied by the second fuel amount portion QFE2.
The purge fuel amount QFP is calculated based on the following formula E11 using the NOx occlusion amount NOXst0 when the purge process should be started.

Here, MNOX represents the molecular weight of NOx, RHC represents the stoichiometric ratio of HC to NOx (for example, 2.5), and MHC represents the average molecular weight of HC (for example, 44).
That is, the purge fuel amount QFP is set to the stoichiometric amount of the NOx occlusion amount NOxst0 with respect to NOx, that is, the amount theoretically required to reduce NOx by NOXst0. This is because the purge process is performed when the inflowing exhaust gas amount is small, so that the residence time of the HC or the reducing agent in the NOx occlusion reduction catalyst 25 becomes considerably long. This is because it is considered to react with the stoichiometric ratio RHC.
As described above, the second fuel amount portion QFE2 represents the amount of fuel that can substantially act as a reducing agent. Therefore, by setting the purge fuel amount QFP in this way, almost all NOx in the NOx storage reduction catalyst 25 can be reduced by the purge process.
The second fuel amount portion QFE2 at the time of the purge process is set as follows.
Part of the additional fuel A once adheres to the exhaust passage upstream of the NOx storage reduction catalyst 25, for example, the exhaust manifold 5, the exhaust turbine 7t, and the inner wall surface of the exhaust pipe 21, and then evaporates and then flows into the NOx storage reduction catalyst 25 There is a case. However, if the additional fuel A once adhering to the inner wall surface of the exhaust passage is referred to as adhering fuel, the adhering fuel flows into the NOx storage reduction catalyst 25 with a delay from the normal timing, which is not preferable. On the other hand, the amount of attached fuel increases as the temperature of the inner wall surface of the exhaust passage upstream of the NOx storage reduction catalyst 25 decreases. Further, the inner wall surface temperature of the exhaust passage upstream of the NOx storage reduction catalyst 25 is expressed by the temperature of the exhaust gas flowing into the NOx storage reduction catalyst 25, for example.
Therefore, in the embodiment according to the present invention, as shown in FIG. 16, the purge is performed so that the temperature Tex of the exhaust gas flowing into the NOx storage reduction catalyst 25 detected by the temperature sensor 26 (FIG. 1) decreases as the temperature Tex decreases. A second fuel amount portion QFE2 at the time of processing is set. As a result, the supply delay of the additional fuel A during the purge process can be suppressed. Note that the second fuel amount portion QFE2 during the purge process is stored in the ROM 42 in the form of a map shown in FIG. 16 as a function of the inflow exhaust gas temperature Tex.
The first fuel portion QFE1 is for reducing the air-fuel ratio AFI of the inflowing exhaust gas from the lean air-fuel ratio obtained by the operating fuel amount QFO to the stoichiometric air-fuel ratio AFS as described above. Therefore, the first fuel amount portion QFE1 is calculated based on the following equation E12.

As described above, the purge process is prohibited when it is determined that the purge condition is not satisfied. In the embodiment according to the present invention, the purge process is also prohibited in the following cases.
If the purge time tP, which is the time required from the start of the purge process to completion, becomes longer, the possibility that the purge process will fail increases.
Therefore, in the embodiment according to the present invention, the predicted purge time PtP that is the predicted value of the purge time tP when it is assumed that the purge process has been performed is obtained, and when the predicted purge time PtP is longer than the predetermined upper limit time Ut, The purge process is prohibited. As a result, the possibility that the purge process will fail can be kept low.
Whether the purge process has succeeded or failed is determined as follows. In other words, if the engine acceleration operation is performed during the purge process, the intake air amount Ga representing the inflow exhaust gas amount greatly increases, and the air-fuel ratio AFI of the inflow exhaust gas cannot be maintained rich. In this case, almost all NOx in the NOx occlusion reduction catalyst 25 cannot be released and reduced. Therefore, when the acceleration operation is performed during the purge process, it is determined that the purge process has failed. On the other hand, if the acceleration operation is not performed during the purge process, the air-fuel ratio AFI of the inflowing exhaust gas is maintained rich. Therefore, when the acceleration operation is not performed during the purge process, it is determined that the purge process is successful.
The predicted purge time PtP is calculated based on, for example, the purge fuel amount QFP, the second fuel amount portion QFE2, and the engine speed. That is, the purge time tP is a time required for the second fuel amount partial integrated value SQFE2 to reach the purge fuel amount QFP. Further, since the second fuel amount portion QFE2 is supplied every expansion stroke or exhaust stroke (see FIG. 7 (i)), the time interval during which the second fuel amount portion QFE2 is supplied is determined according to the engine speed. . Therefore, the predicted purge time PtP can be predicted based on the purge fuel amount QFP, the second fuel amount portion QFE2, and the engine speed.
On the other hand, when the purge process is performed, the excess fuel ratio FER (E10) increases. It is not preferable that the excess fuel ratio FER increases.
Therefore, in the embodiment according to the present invention, the predicted excess fuel ratio PFER, which is the predicted value of the excess fuel ratio FER when it is assumed that the purge process is performed, is obtained, and the predicted excess fuel ratio PFER is set to a predetermined upper limit value UR2. When exceeding, purge processing is prohibited. As a result, it is possible to prevent the excess fuel ratio FER from increasing. The upper limit value UR2 may be the same as or different from the above upper limit value UR1.
The predicted fuel excess ratio PFER is calculated as follows, for example. That is, the purge fuel amount QFP is calculated from the current NOx occlusion amount NOXst, the predicted purge time PtP is calculated from the purge fuel amount QFP, and the operating fuel amount QFO total value and the first fuel amount partial total value at the predicted purge time PtP are calculated. Each is calculated. Then, the operating fuel amount QFO total value is added to the current operating fuel amount integrated value SQFO, and the first fuel amount partial total value and the purge fuel amount QFP are added to the excess fuel amount integrated value SQEF. Next, the excess fuel ratio FER (= SQFE / SQFO) calculated from the equation E10 represents the predicted excess fuel ratio PFER.
When the purge process is performed, the NOx occlusion amount NOXst is updated as follows.
That is, as described above, the purge process may succeed or may fail. When the purge process is successful, it can be considered that all of the NOx in the NOx storage reduction catalyst 25 has been released and reduced. However, this cannot be considered when the purge process fails.
Therefore, in the embodiment according to the present invention, it is determined whether the purge process is successful or unsuccessful, and when it is determined that the purge process is successful, the NOx occlusion amount NOXst is returned to zero. On the other hand, when it is determined that the purge process has failed, the NOx occlusion amount NOXst is held at the value at the start of the purge process. As a result, the NOx occlusion amount NOXst is prevented from greatly deviating from the actual NOx occlusion amount.
In summary, as shown in FIG. 17B, when the acceleration operation is performed during the purge process, it is determined that the purge process has failed, and the NOx occlusion amount NOXst is held at the value at the start of the purge process. On the other hand, as shown in FIG. 17A, when the acceleration operation is not performed during the purge process, it is determined that the purge process is successful, and the NOx occlusion amount NOXst is returned to zero.
FIG. 18 shows an exhaust purification control routine of the embodiment according to the present invention. This routine is executed by interruption every predetermined time.
Referring to FIG. 18, in step 100, it is determined whether or not the rich flag XR is set. The rich flag XR is set when the rich process is to be executed (XR = 1), and is reset otherwise (XR = 0). When the rich flag XR is set, the routine proceeds to step 101 where a rich processing routine is executed. This rich processing routine is shown in FIG. On the other hand, when the rich flag XR is reset, the routine proceeds from step 100 to step 102, where it is determined whether or not the purge flag XP is set. The purge flag XP is set when the purge process is to be executed (XP = 1), and is reset otherwise (XP = 0). When the purge flag XP is set, the routine proceeds to step 103 where a purge process routine is executed. This purge processing routine is shown in FIG. In contrast, when the purge flag XP is reset, the processing cycle is terminated.
Referring to FIG. 19 showing the rich processing routine, in step 110, the NOx occlusion amount NOXst calculated in the routines of FIGS. 24 and 25 is read. In the subsequent step 111, the NOx occlusion speed Vst of the NOx occlusion reduction catalyst 25 is calculated from the NOx occlusion amount NOXst using the map of FIG. In the following step 112, it is determined whether or not the NOx storage speed Vst is equal to or higher than the target NOx storage speed TVst. When Vst <TVst, the routine proceeds to step 113 where the excess fuel ratio FER calculated in the routine of FIG. 28 is read. In the following step 114, it is determined whether or not the excess fuel ratio FER is equal to or lower than the upper limit value UR1. When FER ≦ UR1, the routine proceeds to step 115 where rich combustion is started or continued.
On the other hand, when Vst <TVst at step 112 or when FER> UR1 at step 114, the routine proceeds to step 116, where rich combustion is stopped, and therefore the rich process is interrupted. Next, the routine proceeds to step 117, where the rich flag XR is reset.
Referring to FIG. 20 showing the purge processing routine, it is determined in step 120 whether or not the purge fuel amount QFP has already been calculated. When the purge fuel amount QFP has not been calculated yet, the routine proceeds to step 121, where the purge fuel amount QFP is calculated using Equation E11. Next, the routine proceeds to step 122. When the purge fuel amount QFP has already been calculated, the routine proceeds from step 120 to step 122. In the following step 122, the first fuel amount portion QFE1 is calculated using the equation E12. In the following step 123, the second fuel amount portion QFE2 is calculated using the map of FIG. In the following step 124, the excess fuel amount QFE is calculated (QFE = QFE1 + QFE2), and the additional fuel A is supplied by the excess fuel amount QFE in the expansion stroke or the exhaust stroke. In the following step 125, the second fuel amount partial integrated value SQFE2 is calculated (SQFE2 = SQFE2 + QFE2). In the following step 126, it is determined whether or not the second fuel amount partial integrated value SQFE2 has reached the purge fuel amount QFP. When SQFE2 <QFP, the processing cycle is terminated, and thus the purge process is continued. In contrast, when SQFE2 ≧ QFP, the routine proceeds from step 126 to step 127, where the purge flag XP is reset. Therefore, the purge process is completed. In the following step 128, the second fuel amount partial integration value SQFE2 is cleared (SQFE2 = 0).
FIG. 21 shows a control routine for the rich flag XR. This routine is executed by interruption every predetermined time.
Referring to FIG. 21, in step 130, it is determined whether or not both the rich flag XR and the purge flag XP are reset. When the rich flag XR or the purge flag XP is set, the processing cycle is terminated. Therefore, the rich process is not started when the rich process or the purge process is being performed. When both the rich flag XR and the purge flag XP are reset, the routine proceeds to step 131, where the excess fuel ratio FER calculated in the routine of FIG. 28 is read. In the following step 132, it is determined whether or not the excess fuel ratio FER is equal to or higher than the lower limit value LR1. When FER ≧ LR1, the routine proceeds to step 133, where the rich process request index IDX calculation routine is executed. This routine is shown in FIG. In the following step 134, it is determined whether or not the rich process request index IDX is larger than the threshold value TH. When IDX> TH, the routine proceeds to step 135 where the rich flag XR is set. Therefore, rich processing is performed.
In contrast, when IDX ≦ TH, the routine proceeds from step 134 to step 136, where the rich flag XR is reset. That is, in this case, rich processing is not performed.
On the other hand, when FER <LR1 at step 132, the routine jumps to step 135 and the rich flag XR is set. That is, in this case, rich processing is performed regardless of the rich processing request index IDX.
22 and 23 show the control routine of the purge flag XR. This routine is executed by interruption every predetermined time.
Referring to FIGS. 22 and 23, in step 140, it is determined whether or not both the rich flag XR and the purge flag XP are reset. When the rich flag XR or the purge flag XP is set, the processing cycle is terminated. Therefore, the purge process is not started when the rich process or the purge process is being performed. When both the rich flag XR and the purge flag XP are reset, the routine proceeds to step 141 where it is determined whether or not the intake air amount Ga is smaller than the upper limit amount UG2. When Ga <UG2, the routine proceeds to step 142 where it is determined whether the catalyst temperature T is equal to or higher than the lower limit temperature LT2. Next, when T ≧ LT2, the routine proceeds to step 143, where it is judged if the NOx occlusion amount NOXst is equal to or larger than the lower limit amount LN2. When NOXst ≧ LN2, the routine then proceeds to step 144 where a routine for calculating the predicted purge time PtP is executed. This routine is illustrated in FIG. In the following step 145, it is determined whether or not the predicted purge time PtP is less than or equal to the upper limit time Ut. When PtP ≦ Ut, the routine proceeds to step 146, where the predicted excess fuel ratio PFER is calculated. In the following step 147, it is determined whether or not the predicted excess fuel ratio PFER is equal to or less than the upper limit value UR2. When PFER ≦ UR2, the routine proceeds to step 148, where the purge flag XP is set. Accordingly, in this case, a purge process is performed.
On the other hand, when Ga ≧ UG2 at step 141, T <LT2 at step 142, NOXst <LN2 at step 143, PtP> UtP at step 145, PFER> UR2 at step 147, then Proceeding to step 149, the purge flag XP is reset. Therefore, the purge process is not performed in this case.
Next, referring to FIG. 24 and FIG. 25 showing the routine for calculating the NOx occlusion amount NOXst, in step 150, it is judged if the purge flag XP is reset. When the purge flag XP is set, that is, when the purge process is being performed, the processing cycle is terminated. That is, the NOx occlusion amount NOXst is not updated during the purge process. In contrast, when the purge flag XP is reset, that is, when the purge process is not performed, the routine proceeds to step 151 where it is determined whether or not the purge flag XP was set in the previous process cycle. When the purge flag XP has been reset in the previous processing cycle, that is, not immediately after the purge processing is completed, the routine proceeds to step 152 where the outflow exhaust gas air-fuel ratio AFO is equal to or lower than the stoichiometric air-fuel ratio AFS, that is, outflow exhaust gas. It is determined whether the air-fuel ratio AFO is the stoichiometric air-fuel ratio or rich. When AFO> AFS, that is, when the air-fuel ratio AFO of the outflow exhaust gas is lean, the routine proceeds to step 153, where the inflow NOx amount is calculated, and the increase NOXin is set to the inflow NOx amount. In the following step 154, the decrease NOXrd is made zero. Next, the routine proceeds to step 157.
On the other hand, when AFO ≦ AFS, that is, when the air-fuel ratio AFO of the outflow exhaust gas is the stoichiometric air-fuel ratio or rich, the routine proceeds to step 155, where the increment NOXin is made zero. In the subsequent step 156, a routine for calculating the decrease NOXrd is executed. This routine is illustrated in FIG. Next, the routine proceeds to step 157.
In step 157, the NOx occlusion amount NOXst is updated using the equation E1.
On the other hand, when the purge flag XP is set in the previous processing cycle, that is, immediately after the purge processing is completed, the routine proceeds from step 151 to step 158 to determine whether or not the purge failure flag XPF is reset. . This purge failure flag XPF is set when it is determined that the purge process has failed (XPF = 1), and otherwise it is reset (XPF = 0). When the purge failure flag XPF is reset, that is, when it is determined that the purge process is successful, the routine proceeds to step 159, where the NOx occlusion amount NOXst is returned to zero.
On the other hand, when the purge failure flag XPF is set, that is, when it is determined that the purge process has failed, the routine proceeds from step 158 to step 160, where the purge failure flag XPF is reset. The processing cycle is then terminated. Therefore, when the purge process fails, the NOx occlusion amount NOXst is not updated, that is, held at the value at the start of the purge process.
Referring to FIG. 26 showing the routine for calculating the decrease NOXrd, in step 170, the air-fuel ratio AFI of the inflowing exhaust gas is read. In the subsequent step 171, the NOx occlusion amount NOXst is read. In the following step 172, the catalyst temperature T is read. In the subsequent step 173, the first correction coefficient KC1, the second correction coefficient KC2, and the third correction coefficient KC3 are determined using the maps of FIGS. 3B, 4B, and 5B. Each is calculated, and the correction coefficient KC is calculated using Formula E6. In the following step 174, the decrease NOXrd is calculated using the equation E2.
FIG. 27 shows a control routine for the purge failure flag XPF. This routine is executed by interruption every predetermined time.
Referring to FIG. 27, in step 180, it is determined whether or not the purge failure flag XPF is reset. When the purge failure flag XPF is reset, the routine proceeds to step 181 where it is determined whether or not the purge flag XP is set. When the purge flag XP is set, that is, during the purge process, the routine proceeds to step 182 where it is determined whether or not an acceleration operation is being performed. When the acceleration operation is being performed, the routine proceeds to step 183 where the purge failure flag XPF is set. In contrast, when the purge failure flag XPF is set at step 180, when the purge flag XP is reset at step 181, or when the acceleration operation is not performed at step 182, the processing cycle is terminated.
FIG. 28 shows a routine for calculating the excess fuel ratio FER. This routine is executed by interruption every predetermined time.
Referring to FIG. 28, in step 190, the operating fuel amount integrated value SQFO is calculated by integrating the operating fuel amount QFO. In the subsequent step 191, the excess fuel amount integrated value SQFE is calculated by integrating the excess fuel amount QFE. In the following step 192, the excess fuel ratio FER is calculated using Formula E10.
FIG. 29 shows a routine for calculating the rich process request index IDX. This routine is executed by interruption every predetermined time.
Referring to FIG. 29, in step 200, it is determined whether or not both the rich flag XR and the purge flag XP are reset. When the rich flag XR or the purge flag XP is set, the processing cycle is terminated. When both the rich flag XR and the purge flag XP are reset, the routine proceeds to step 201, where the index IDXeg is calculated using the map of FIG. In the subsequent step 202, the index IDXet is calculated using the map of FIG. In the following step 203, the first index IDX1 is calculated using the mathematical formula E8. In the following step 204, the index IDXsv is calculated using the map of FIG. In the following step 205, the index IDXsp is calculated using the map of FIG. In the following step 206, the second index IDX2 is calculated using the mathematical formula E9. In the subsequent step 207, the rich process request index IDX is calculated using the mathematical formula E7.
FIG. 30 shows another embodiment of the compression ignition type internal combustion engine. In the embodiment shown in FIG. 30, the reducing agent addition valve 31 is attached to the exhaust passage upstream of the NOx storage reduction catalyst 25, for example, the exhaust manifold 5. When the rich process or the purge process is to be performed, a reducing agent such as CO or HC is added in a pulse form from the reducing agent addition valve 31, thereby making the air-fuel ratio of the inflowing exhaust gas temporarily rich.
Next, another embodiment of the operation for calculating the NOx occlusion amount NOXst will be described.
In the embodiments according to the present invention described so far, when it is determined that the purge process has failed, the NOx occlusion amount NOXst is held at the value at the start of the purge process. This means that NOx is not released from the NOx storage reduction catalyst 25 when it is determined that the purge process has failed. However, the air-fuel ratio of the inflowing exhaust gas is maintained rich until the acceleration operation is performed, and NOx is released from the NOx storage reduction catalyst 25 and reduced.
Therefore, in another embodiment according to the present invention, when it is determined that the purge process has failed, the NOx occlusion amount NOXst is decreased by the purge process, and the NOx occlusion amount NOXst is decreased by the decrease by the purge process. . As a result, the NOx occlusion amount NOXst can be accurately maintained even when it is determined that the purge process has failed.
That is, when it is determined that the purge process has failed, the NOx occlusion amount NOXst is updated based on the following formula E13.

Here, PNrd represents a decrease amount of the NOx occlusion amount NOXst by one purge process.
The amount of decrease PNrd due to this purge process is calculated as follows, for example. That is, as described above, in the purge process until the acceleration operation is performed, it is considered that HC or the reducing agent and NOx react with the stoichiometric ratio RHC. Therefore, if the effective HC amount, which is the amount of HC that effectively acts on the release and reduction of NOx during the purge process, is known, the decrease PNrd due to the purge process can be determined.
On the other hand, the effective HC amount is represented by the air-fuel ratio AFI of the inflowing exhaust gas during the purge process. Specifically, the effective HC amount increases as the deviation (AFS-AFI) of the air-fuel ratio AFI of the inflowing exhaust gas with respect to the stoichiometric air-fuel ratio AFS increases. In this case, the area of the closed region surrounded by the straight line representing the stoichiometric air-fuel ratio AFS and the curve representing the inflowing exhaust gas AFI represents the total effective HC amount during the purge process.
Therefore, in another embodiment according to the present invention, the air-fuel ratio AFI of the inflowing exhaust gas is stored in the RAM 43 during the purge process, and when it is determined that the purge process has failed, the stored air-fuel ratio of the inflowing exhaust gas is stored. Based on the AFI, the area of the closed region is calculated. Next, the entire effective HC amount is calculated based on this area, and the decrease amount PNrd due to the purge process is calculated based on the entire effective HC amount. Next, the NOx occlusion amount NOXst is updated using Equation 13.
In summary, when it is determined that the purge process has failed as shown in FIG. 31, the decrease amount PNrd due to the purge process is calculated based on the area of the closed region AR, and the NOx occlusion amount NOXst is decreased by the decrease amount PNrd. The
32 and 33 show another embodiment of the routine for calculating the NOx occlusion amount NOXst. This routine differs from the routine shown in FIGS. 24 and 25 in the following points.
That is, when it is determined that the purge process has failed, after the purge failure flag XPF is reset in step 160, the process proceeds to step 161, and a decrease amount PNrd due to the purge process is calculated. In the following step 162, the NOx occlusion amount NOXst is updated using the equation E13.
As described above, HC or reducing agent in the inflowing exhaust gas may be consumed by the three-way catalyst 24. Therefore, the reduction amount PNrd due to the purge process may be corrected based on the amount of reducing agent consumed by the three-way catalyst 24 during the purge process.
Incidentally, when the rich process is performed, the NOx occlusion amount NOXst decreases as shown in FIG. When the decrease amount of the NOx occlusion amount NOXrd by one rich process is represented by RNrd, in the example shown in FIG. 34, the NOx occlusion amount NOXst decreases by the decrease amount RNrd from the NOx occlusion amount NOXst0 at the start of the rich process.
The amount of decrease RNrd due to this rich process can be estimated as follows. That is, the reduction amount RNrd due to the rich process is equal to the sum of the reduction amount NOXrd. Then, the following formula E14 is obtained by integrating the formula E3.

Assuming that the inflow CO amount QCO is constant during the rich process, the integral term of the equation E14 is equal to the area of the trapezoid hatched in FIG. 34, and is represented by the following equation E15.

Here, tR is the time during which the air-fuel ratio of the inflowing exhaust gas is rich due to the rich process.
Therefore, the reduction amount RNrd due to the rich process is expressed by the following equation E16 from the equations E14 and E15.

Here, Kt is expressed by the following formula E17.

Therefore, if the rich time tR is given, the reduction amount RNrd due to the rich process can be estimated using the mathematical formula E16. As a result, the reduction amount RNrd due to the rich process can be estimated easily and accurately.
On the other hand, the utilization efficiency EFF of CO or the reducing agent when the rich process is performed is expressed by, for example, the following formula E18.

Here, QFR represents a rich fuel amount that is an integrated value of the second fuel amount portion QFE2 in one rich process, and can be obtained based on the rich time tR and the target rich air-fuel ratio.
Then, if the reduction amount RNrd due to the rich process is estimated by the equation E16, the utilization efficiency EFF can be estimated by the equation E18. In this way, it is possible to accurately and easily estimate the utilization efficiency EFF without actually performing rich processing.
Therefore, in general terms, the reduction amount RNrd of the NOx occlusion amount due to the rich processing is the NOx occlusion amount NOXst0 at the start of the rich processing, the inflow reducing agent amount QCO and the catalyst temperature T that are assumed to be constant during the rich processing. It will be estimated based on.

DESCRIPTION OF SYMBOLS 1 Engine body 2 Combustion chamber 3 Fuel injection valve 21 Exhaust pipe 25 NOx storage reduction catalyst 27 Air-fuel ratio sensor 28 Temperature sensor

Claims

An exhaust purification device for an internal combustion engine having an exhaust passage,
A NOx occlusion reduction catalyst disposed in the exhaust passage, which stores NOx contained in the inflowing exhaust gas when the air-fuel ratio of the inflowing exhaust gas is lean and stores the NOx occluded when the air-fuel ratio of the inflowing exhaust gas becomes rich A NOx occlusion reduction catalyst that releases and reduces;
A rich processing unit that performs rich processing for temporarily switching the air-fuel ratio of the inflowing exhaust gas to be rich in order to partially release and reduce NOx stored in the NOx storage reduction catalyst;
A NOx occlusion amount calculation unit for calculating the NOx occlusion amount of the NOx occlusion reduction catalyst;
A rich process control unit for controlling the rich process based on the NOx occlusion amount;
Comprising
At the time of the rich processing, the reduction amount per unit time of the NOx occlusion amount is based on the inflow reducing agent amount that is the amount of reducing agent in the inflowing exhaust gas, the NOx occlusion amount, and the catalyst temperature that is the temperature of the NOx occlusion reduction catalyst An exhaust purification device for an internal combustion engine calculated.
The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the decrease is calculated without obtaining the amount of NOx flowing out from the NOx occlusion reduction catalyst.
The decrease is

Here, QRED is calculated based on the inflow reducing agent amount, NOXst is the NOx occlusion amount, T is the catalyst temperature, E is the activation energy, and R is the gas constant. The exhaust emission control device for an internal combustion engine according to claim 1.
The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the reducing agent is made of carbon monoxide, hydrogen, or hydrocarbon.
2. The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the decrease is corrected based on a reducing agent passage amount that is a reducing agent amount that passes through the NOx occlusion reduction catalyst without reducing NOx during the rich processing.
The exhaust gas purification apparatus for an internal combustion engine according to claim 5, wherein the reducing agent passage amount is represented by an inflowing exhaust gas amount.
The exhaust gas purification apparatus for an internal combustion engine according to claim 5, wherein the reducing agent passage amount is represented by a degree of clogging of a casing in which the NOx storage reduction catalyst is accommodated.
The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the decrease is corrected based on a degree of deterioration of the NOx storage reduction catalyst.
The exhaust purification device for an internal combustion engine according to claim 1, wherein the amount of the inflow reducing agent is obtained based on an air-fuel ratio of the inflowing exhaust gas.
When the air-fuel ratio of the inflowing exhaust gas is lean, an oxygen storage catalyst that stores oxygen contained in the exhaust gas and releases the stored oxygen when the air-fuel ratio of the inflowing exhaust gas becomes rich is an exhaust passage upstream of the NOx storage reduction catalyst 2. The internal combustion engine according to claim 1, wherein the inflow reducing agent amount is corrected based on a reducing agent consumption amount that is a reducing agent amount in the exhaust gas that is disposed in the exhaust gas and is consumed by the oxygen storage catalyst during the rich processing. Exhaust purification equipment.
The exhaust gas purification device for an internal combustion engine according to claim 10, wherein the amount of consumption of the reducing agent is represented by an intake air amount.
The exhaust gas purification apparatus for an internal combustion engine according to claim 10, wherein the reducing agent consumption is represented by an oxygen storage amount of an oxygen storage catalyst when the rich process is started.
13. The exhaust gas purification apparatus for an internal combustion engine according to claim 12, wherein the oxygen storage amount is represented by a time during which the air-fuel ratio of the inflowing exhaust gas is kept lean until the rich process is started.
2. The purge processing unit according to claim 1, further comprising a purge processing unit that performs a purge process that temporarily changes the air-fuel ratio of the inflowing exhaust gas to be rich in order to release and reduce almost all of the NOx stored in the NOx storage reduction catalyst. Exhaust gas purification device for internal combustion engine.
Purge when the inflow exhaust gas amount is higher than a predetermined upper limit amount, the catalyst temperature is lower than a predetermined lower limit temperature, or the NOx occlusion amount is lower than a predetermined lower limit amount The exhaust emission control device for an internal combustion engine according to claim 14, wherein the processing is prohibited.
The purge process is performed when the inflowing exhaust gas amount is less than a predetermined upper limit amount, the catalyst temperature is higher than a predetermined lower limit temperature, and the NOx occlusion amount is larger than a predetermined lower limit amount. The exhaust emission control device for an internal combustion engine according to claim 14, wherein the exhaust gas purification device is performed.
Of the amount of reducing agent supplied during the purge process, the integrated value of the reducing agent amount portion for making the air-fuel ratio of the inflowing exhaust gas rich beyond the theoretical air-fuel ratio is set to the stoichiometric amount of the NOx occlusion amount with respect to NOx. The exhaust emission control device for an internal combustion engine according to claim 14.
15. The exhaust gas purification apparatus for an internal combustion engine according to claim 14, wherein it is determined whether the purge process is successful or unsuccessful, and the NOx occlusion amount is returned to zero when it is determined that the purge process is successful.
15. The exhaust gas purification apparatus for an internal combustion engine according to claim 14, wherein it is determined whether the purge process has succeeded or failed, and when it is determined that the purge process has failed, the NOx occlusion amount is held at a value at the start of the purge process.
It is determined whether the purge process has succeeded or failed. When it is determined that the purge process has failed, a decrease in the NOx occlusion amount due to the purge process is obtained based on the air-fuel ratio of the inflowing exhaust gas during the purge process and The exhaust gas purification apparatus for an internal combustion engine according to claim 14, wherein the NOx occlusion amount is reduced by a reduction amount by the purge process.
Of the amount of reducing agent supplied during the purge process, the reducing agent amount portion for making the air-fuel ratio of the inflowing exhaust gas rich beyond the stoichiometric air-fuel ratio is based on the inner wall surface temperature of the exhaust passage upstream of the NOx storage reduction catalyst. The exhaust emission control device for an internal combustion engine according to claim 14, which is set.
When it is assumed that the purge process has been performed, the purge time, which is the time required for the purge process, is predicted based on the reducing agent amount portion, and when the predicted purge time is longer than a predetermined upper limit time, the purge is performed. The exhaust emission control device for an internal combustion engine according to claim 21, wherein the processing is prohibited.
The first index determined according to the reducing agent utilization efficiency when it is assumed that the rich process is performed is determined, and it is determined whether or not the rich process should be performed based on the first index. An exhaust gas purification apparatus for an internal combustion engine as described.
24. The exhaust gas purification apparatus for an internal combustion engine according to claim 23, wherein the first index is obtained based on one or both of an inflowing exhaust gas amount and the catalyst temperature.
2. The internal combustion engine according to claim 1, wherein a second index determined according to the NOx storage capacity of the NOx storage reduction catalyst is obtained, and it is determined based on the second index whether rich processing should be performed. Exhaust purification device.
The second index is one or both of the NOx occlusion speed of the NOx occlusion reduction catalyst and the NOx passage amount that is the NOx occlusion amount that passes through the NOx occlusion reduction catalyst when the air-fuel ratio of the inflowing exhaust gas is lean. The exhaust emission control device for an internal combustion engine according to claim 25, which is obtained based on the above.
A first index determined according to the reducing agent utilization efficiency when it is assumed that the rich process has been performed, and a second index determined according to the NOx storage capacity of the NOx storage reduction catalyst are obtained. The exhaust emission control device for an internal combustion engine according to claim 1, wherein whether or not to perform rich processing is determined based on the index and the second index.
The first index is obtained based on the inflow exhaust gas amount and the catalyst temperature, and the second index is NOx when the NOx occlusion speed of the NOx storage reduction catalyst and the air-fuel ratio of the inflow exhaust gas are lean. 28. The exhaust gas purification apparatus for an internal combustion engine according to claim 27, obtained based on an NOx passage amount that is an NOx amount passing through the storage reduction catalyst.
The ratio of the integrated value of the amount of reducing agent supplied to make the air-fuel ratio of the inflowing exhaust gas rich with respect to the integrated value of the fuel amount supplied for engine operation is determined, and the ratio is a predetermined lower limit. The exhaust emission control device for an internal combustion engine according to claim 1, wherein the rich process is performed when the value becomes smaller than the value.
A ratio of the integrated value of the reducing agent supplied to make the air-fuel ratio of the inflowing exhaust gas rich with respect to the integrated value of the fuel amount supplied for engine operation is obtained, and the ratio is determined in advance during the rich processing. 2. The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the rich processing is interrupted when the set upper limit value is exceeded.
When it is assumed that purge processing has been performed with the ratio of the integrated value of the reducing agent supplied to make the air-fuel ratio of the inflowing exhaust gas rich relative to the integrated value of the fuel amount supplied for engine operation 15. The exhaust gas purification apparatus for an internal combustion engine according to claim 14, wherein the purge process is prohibited when the ratio is predicted and the predicted ratio exceeds a predetermined upper limit value.
2. The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the rich process is completed when the NOx occlusion speed of the NOx occlusion reduction catalyst determined according to the NOx occlusion amount reaches a predetermined target value during the rich process.
The reduction amount of the NOx occlusion amount due to the rich processing is estimated based on the NOx occlusion amount at the start of the rich processing, and the inflow reducing agent amount and the catalyst temperature that are assumed to be constant during the rich processing. An exhaust gas purification apparatus for an internal combustion engine as described.
The exhaust purification device for an internal combustion engine according to claim 1, wherein the rich processing is performed by temporarily performing combustion under a rich air-fuel ratio.
2. The exhaust gas purification apparatus for an internal combustion engine according to claim 1, wherein the reduction amount is repeatedly calculated during the rich processing, and the NOx occlusion amount is repeatedly updated by the calculated reduction amount.