JP3508192B2 - Exhaust gas purification device for internal combustion engine - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exhaust gas purification device for an internal combustion engine.

[0002]

2. Description of the Related Art When the air-fuel ratio of the exhaust gas that flows in is lean,
No when_xAnd the air-fuel ratio of the inflowing exhaust gas is
The absorbed NO when the air-fuel ratio or rich_xEmit
NO _xAbsorbent is placed in the engine exhaust passage for lean mixing
NO when the air is burned_xNO_xFor absorbent
Absorb and burn a stoichiometric air-fuel ratio mixture or rich mixture.
NO when tightened_xAbsorbent to NO_xEmit
NO released with_xInternal combustion engine designed to reduce
Is known (see International Publication WO93 / 07363).
See).

Further, when the air-fuel ratio of the air-fuel mixture is changed from the lean air-fuel ratio to the stoichiometric air-fuel ratio or from the stoichiometric air-fuel ratio to the lean air-fuel ratio, the air-fuel ratio is gradually decreased or gradually increased according to the operating state of the engine. As a result, an internal combustion engine is known in which the engine output torque is prevented from changing suddenly when the air-fuel ratio is switched (see Japanese Patent Laid-Open No. 59-7741).

[0004]

However, in the internal combustion engine in which the NO _x absorbent is arranged in the engine exhaust passage, the air-fuel ratio of the air-fuel mixture is changed from lean air-fuel ratio to stoichiometric air-fuel ratio or from stoichiometric air-fuel ratio to lean air-fuel ratio. When the air-fuel ratio is gradually decreased or gradually increased in order to prevent a sudden change in the engine output torque during switching, a large amount of NO _x is released into the atmosphere without being absorbed by the NO _x absorbent.

That is, if the chain line in FIG. 19 is the boundary line for switching between the lean air-fuel ratio and the stoichiometric air-fuel ratio, the throttle opening is slightly increased to slightly increase the vehicle speed, and the engine load exceeds the boundary line. The air-fuel ratio of the mixture is gradually changed from the lean air-fuel ratio to the stoichiometric air-fuel ratio. However, when the air-fuel ratio of the air-fuel mixture is switched from the lean air-fuel ratio to the stoichiometric air-fuel ratio, the engine output is greatly increased, so that the vehicle speed increases more than the driver expected. As a result, this time the throttle opening is reduced to reduce the vehicle speed.

When the throttle opening is made smaller and the engine load becomes lower than the boundary line, the air-fuel ratio of the air-fuel mixture is gradually changed from the stoichiometric air-fuel ratio to the lean air-fuel ratio. However, when the air-fuel ratio of the air-fuel mixture is switched from the stoichiometric air-fuel ratio to the lean air-fuel ratio, the engine output is greatly reduced, and the vehicle speed is reduced more than the driver expected. As a result, this time, the throttle opening is increased to increase the vehicle speed. In this way, the opening operation and the closing operation of the throttle valve are alternately repeated, and the air-fuel ratio fluctuates between the lean air-fuel ratio and the stoichiometric air-fuel ratio.

[0007] However NO NO _x absorption rate of _x absorbent is low in the slightly lean region than the stoichiometric air-fuel ratio, also in large quantities from the engine is slightly lean region than the stoichiometric air-fuel ratio NO _x
Is released. As a result, in a region slightly leaner than the stoichiometric air-fuel ratio, NO _x discharged from the engine cannot be absorbed well by the NO _x absorbent, and a part of NO _x is released to the atmosphere. Therefore, as shown in FIG. 19, when the air-fuel ratio of the air-fuel mixture goes back and forth between the lean air-fuel ratio and the stoichiometric air-fuel ratio, the air-fuel ratio passes through a region slightly leaner than the stoichiometric air-fuel ratio, so that There is a problem that NO _x is discharged into the atmosphere every time. Such a problem similarly occurs when the air-fuel ratio of the air-fuel mixture goes back and forth between the lean air-fuel ratio and the rich air-fuel ratio.

[0008]

In order to solve the above problems, according to the present invention, when the air-fuel ratio of the inflowing exhaust gas is lean, NO _x is absorbed, and the air-fuel ratio of the inflowing exhaust gas is the theoretical air-fuel ratio. when the fuel ratio or rich is arranged to _the NO _x absorbent to release the absorbed NO _x in the engine exhaust passage, and the theoretical air-fuel ratio or rich combustion region of the lean burn region and a high load side of the operating region of the engine low load side In the lean combustion region, the lean air-fuel mixture is burned to absorb the NO _x discharged from the engine in the NO _x absorbent, and in the theoretical air-fuel ratio or rich combustion region, the air-fuel ratio mixture or rich mixture is added. in an internal combustion engine which is adapted to release _the NO _x absorbed in _the NO _x absorbent by burned gas,
When the operating region of the engine shifts from the lean combustion region to the stoichiometric air-fuel ratio combustion region, the air-fuel ratio of the air-fuel mixture is gradually reduced.
Together with the engine operating range lean from the theoretical air-fuel ratio combustion range
When shifting to the combustion region, the air-fuel ratio of the air-fuel mixture is gradually increased.
Comprising a Kusuru air gradual change means, the lean combustion region and Theory
The boundary line between the air-fuel ratio combustion region is the first boundary line at the transition from the lean combustion region to the stoichiometric air-fuel ratio combustion region, and the first boundary line is the boundary at the transition from the theoretical air-fuel ratio combustion region to the lean combustion region. The second boundary line is set on the low load side with respect to the first boundary line.

Further, according to the present invention, in order to solve the above problems, NO _x is absorbed when the air-fuel ratio of the inflowing exhaust gas is lean, and when the air-fuel ratio of the inflowing exhaust gas is the stoichiometric air-fuel ratio or rich. NO releasing absorbed NO _{x
x An} absorbent is placed in the engine exhaust passage to divide the operating region of the engine into a lean combustion region on the low load side and a stoichiometric air-fuel ratio or rich combustion region on the high load side. NO _x emitted from the engine after burning
Was allowed to absorb into _the NO _x absorbent, when the stoichiometric air-fuel ratio or a rich burn zone is so as to release the NO _x absorbed in _the NO _x absorbent by burned air-fuel mixture or a rich mixture of the theoretical air-fuel ratio internal combustion In the engine, when the operating region of the engine shifts from the lean combustion region to the stoichiometric air-fuel ratio combustion region, the air-fuel ratio of the air-fuel mixture is gradually reduced and the engine
The operating range has changed from the stoichiometric air-fuel ratio combustion range to the lean combustion range.
When gradually increasing the air-fuel ratio of the air-fuel mixture when the engine is operated, the engine operating region shifts from the lean combustion region to the stoichiometric air-fuel ratio combustion region, and then the engine operating region becomes the lean combustion region and the theoretical air-fuel ratio. Estimating means for estimating that there is a possibility of repeating the fuel ratio combustion region alternately, and when it is estimated that the operating region of the engine may alternate between the lean combustion region and the stoichiometric air-fuel ratio combustion region Air-fuel ratio temporarily
And an air-fuel ratio fixing means for fixing the lean air-fuel ratio or the stoichiometric air-fuel ratio .

Further, according to the present invention, in order to solve the above-mentioned problems, the above-mentioned estimation means is such that after the operating region of the engine is changed from the lean combustion region to the stoichiometric air-fuel ratio combustion region, the operating region of the engine is again lean combustion region. It is estimated that there is a possibility that the operating region of the engine may alternate between the lean combustion region and the stoichiometric air-fuel ratio combustion region when the amount of change in the engine load before the transition to is smaller than a predetermined amount. .

[0011]

[0012]

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to FIG. 1, 1 is an engine body, 2 is a piston, 3 is a combustion chamber, 4 is a spark plug, 5 is an intake valve, 6 is an intake port, 7 is an exhaust valve, and 8 is an exhaust port. Show. The intake port 6 is connected to the surge tank 10 via a corresponding branch pipe 9, and each branch pipe 9 is provided with a fuel injection valve 11 for injecting fuel into the intake port 6. The surge tank 10 is connected to an air cleaner 13 via an intake duct 12, and a throttle valve 14 is arranged in the intake duct 12. On the other hand, the exhaust port 8 is connected via an exhaust manifold 15 and an exhaust pipe 16 to a casing 17 containing a NO _x absorbent 18.

The electronic control unit 30 comprises a digital computer, and a ROM (read only memory) 32, a RAM (random access memory) 33, a CPU (microprocessor) 34, an input port 35 which are mutually connected by a bidirectional bus 31. And an output port 36. A pressure sensor 19 for generating an output voltage proportional to the absolute pressure in the surge tank 10 is arranged in the surge tank 10, and the output voltage of the pressure sensor 19 corresponds to A
It is input to the input port 35 via the D converter 37. A throttle opening sensor 20 that generates an output voltage proportional to the throttle opening is attached to the throttle valve 14, and the output voltage of the throttle opening sensor 20 corresponds to the AD.
It is input to the input port 35 via the converter 37. An air-fuel ratio sensor 21 is arranged in the exhaust manifold 15, and the output voltage of this air-fuel ratio sensor 21 corresponds to the AD converter 3
It is input to the input port 35 via 7. Further, the input port 35 is connected to the rotation speed sensor 22 that generates an output pulse representing the engine rotation speed. On the other hand, the output port 36
Are connected to the spark plug 4 and the fuel injection valve 11 via corresponding drive circuits 38, respectively.

In the internal combustion engine shown in FIG. 1, the fuel injection time TAU is calculated based on the following equation, for example. TAU = f · TP · K · FAF where f is a constant, TP is the basic fuel injection time, K is a correction coefficient, and FAF is a feedback correction coefficient. The basic fuel injection time TP indicates the fuel injection time required to make the air-fuel ratio of the air-fuel mixture supplied into the engine cylinder the stoichiometric air-fuel ratio. This basic fuel injection time TP is obtained in advance by experiments and is stored in advance in the ROM 32 in the form of a map as shown in FIG. 2 as a function of the absolute pressure PM in the surge tank 10 and the engine speed N. The correction coefficient K is a coefficient for controlling the air-fuel ratio of the air-fuel mixture supplied to the engine cylinder, and if K = 1.0, the air-fuel mixture supplied to the engine cylinder becomes the stoichiometric air-fuel ratio. On the other hand, K <
When 1.0, the air-fuel ratio of the air-fuel mixture supplied to the engine cylinder becomes larger than the stoichiometric air-fuel ratio, that is, becomes lean, and when K> 1.0, the air-fuel ratio of the air-fuel mixture supplied to the engine cylinder becomes empty. The fuel ratio becomes smaller than the stoichiometric air-fuel ratio, that is, becomes rich.

The feedback correction coefficient FAF is K = 1.
When 0, that is, when the air-fuel ratio of the air-fuel mixture supplied to the engine cylinder should be the stoichiometric air-fuel ratio, a coefficient for accurately matching the air-fuel ratio with the stoichiometric air-fuel ratio based on the output signal of the air-fuel ratio sensor 21. Is. This feedback correction coefficient FAF
Moves up and down about 1.0, and this FAF
Decreases when the mixture becomes rich, and increases when the mixture becomes lean. When K <1.0 or K> 1.0, FAF is fixed at 1.0.

The target air-fuel ratio of the air-fuel mixture to be supplied into the engine cylinder, that is, the value of the correction coefficient K is changed in accordance with the operating state of the engine. In the embodiment of the present invention, it is basically shown in FIG. Thus, it is predetermined as a function of the absolute pressure PM in the surge tank 10 and the engine speed N. That is, as shown in FIG. 3, in the idling operation region on the lower load side than the boundary line Q, K = 1.0, that is, the air-fuel ratio of the air-fuel mixture is set to the stoichiometric air-fuel ratio, and between the boundary lines Q and R. K is less than 1.0 in the low load operation region, that is, the air-fuel ratio of the air-fuel mixture is lean, and K is 1.0 in the high-load operation region between the boundary line R and the boundary line S, that is, the air-fuel ratio of the air-fuel mixture. Is the stoichiometric air-fuel ratio, and K> in the full-load operation region on the higher load side than the boundary line S.
1.0, that is, the air-fuel ratio of the air-fuel mixture is rich.

FIG. 4 schematically shows the concentrations of typical components in the exhaust gas discharged from the combustion chamber 3. As can be seen from FIG. 4, the concentration of unburned HC and CO in the exhaust gas discharged from the combustion chamber 3 increases as the air-fuel ratio of the air-fuel mixture supplied into the combustion chamber 3 becomes richer, and is discharged from the combustion chamber 3. The concentration of oxygen O _{2 in} the generated exhaust gas increases as the air-fuel ratio of the air-fuel mixture supplied into the combustion chamber 3 becomes leaner.

NO stored in the casing 17_x
The absorbent 18 uses, for example, alumina as a carrier, and
For example, potassium K, sodium Na, lithium Li, ce
Alkali metals such as sium Cs, barium Ba, cal
Alkaline earths such as Ca, lanthanum La,
At least one selected from rare earths such as thorium Y
And a noble metal such as platinum Pt. organ
Intake passage and NO_xProvided in the exhaust passage upstream of the absorbent 18.
The ratio of air and fuel (hydrocarbons) fed is NO _xabsorption
When called the air-fuel ratio of the exhaust gas flowing into the agent 18, this NO_x
The absorbent 18 is used when the air-fuel ratio of the inflowing exhaust gas is lean.
NO_xIs absorbed, and the oxygen concentration in the exhaust gas is reduced.
NO absorbed by_xReleases NO_xTo absorb and release
U Note that NO_xFuel in the exhaust passage upstream of the absorbent 18
(Hydrocarbon) or inflow / outflow when air is not supplied
The air-fuel ratio of the air-gas is the air-fuel ratio of the air-fuel mixture supplied into the combustion chamber 3.
The ratio and therefore NO in this case_xAbsorbent 18 burns
When the air-fuel ratio of the air-fuel mixture supplied to the firing chamber 3 is lean
Is NO_xOf the air-fuel mixture in the combustion chamber 3
NO absorbed when oxygen concentration decreases_xTo release
Become.

If the above-mentioned NO _x absorbent 18 is arranged in the exhaust passage of the engine, the NO _x absorbent 18 actually acts to absorb and release NO _x , but the detailed mechanism of this absorbing and releasing effect is not clear. There is also. However, it is considered that this absorbing / releasing action is performed by the mechanism shown in FIG. Next, this mechanism will be described by taking the case where platinum Pt and barium Ba are supported on the carrier as an example, but the same mechanism can be obtained by using other noble metals, alkali metals, alkaline earths and rare earths.

That is, when the air-fuel ratio of the air-fuel mixture supplied into the combustion chamber 3 is made lean and the inflowing exhaust gas becomes lean, the oxygen concentration in the inflowing exhaust gas greatly increases, and as shown in FIG. As shown, these oxygen O ₂ attaches to the surface of platinum Pt in the form of O ₂ ⁻ or O ²⁻ . On the other hand, N in the inflowing exhaust gas
O reacts with O ₂ ⁻ or O ^{2 −} on the surface of platinum Pt, and NO
It becomes ₂ (2NO + O ₂ → 2NO ₂ ). Then, a part of the generated NO ₂ is absorbed on the platinum Pt while being absorbed in the absorbent to bond with barium oxide BaO.
As shown in (A), it diffuses into the absorbent in the form of nitrate ion NO ₃ ⁻ . In this way, NO _x is absorbed in the NO _x absorbent 18.

Platinum as long as the oxygen concentration in the inflowing exhaust gas is high
NO on the surface of Pt₂Is generated and the absorbent NO_xAbsorption capacity
NO unless power is saturated₂Is absorbed in the absorbent and nitric acid
Ionic NO₃ ^-Is generated. On the other hand, the inflow exhaust gas
NO in the oxygen₂When the production of
Reaction is in the opposite direction (NO₃ ^-→ NO₂), And thus suck
Nitrate ion NO in the sorbent₃ ^-Is NO₂From the absorbent in the form of
Is released. That is, the oxygen concentration in the inflowing exhaust gas decreases
And NO_xAbsorbent 18 to NO_xWill be released
It As shown in Fig. 4, the degree of leanness of the inflowing exhaust gas
The lower the value, the lower the oxygen concentration in the inflowing exhaust gas.
Even if the lean degree of the inflowing exhaust gas is lowered by
NO even if the air-fuel ratio of the exhaust gas is lean_xAbsorbent 18
To NO _xWill be released.

On the other hand, at this time, when the air-fuel mixture supplied into the combustion chamber 3 is made rich and the air-fuel ratio of the inflowing exhaust gas becomes rich, a large amount of unburned H is emitted from the engine as shown in FIG.
C and CO are discharged, and these unburned HC and CO are platinum Pt.
It is oxidized by reacting with the oxygen O ₂ ⁻ or O ^{2 −} above.
Further, when the air-fuel ratio of the inflowing exhaust gas becomes rich, the oxygen concentration in the inflowing exhaust gas drops extremely, so
O ₂ is released, and this NO ₂ is reduced by reacting with unburned HC and CO as shown in FIG. 5 (B). When NO ₂ is no longer present on the surface of platinum Pt in this way, NO ₂ is released one after another from the absorbent. Therefore NO _x from the NO _x absorbent 18 in a short time when the air-fuel ratio of the inflowing exhaust gas is made rich, that is released.

That is, the air-fuel ratio of the inflowing exhaust gas is made rich.
Then, first, unburned HC and CO are O on platinum Pt.₂ ^-or
Is O^2-Reacts instantly with and is oxidized, then platinum
O on Pt₂ ^-Or O^2-Is consumed, but unburned HC,
If CO remains, the unburned HC and CO absorb the absorbent.
NO released from_xAnd NO emitted from the engine _x
Is reduced. Therefore, the air-fuel ratio of the inflowing exhaust gas is
No in a short time if you switch_xAbsorbed by absorbent 18
NO_xIs released, and this released NO
_xNO is reduced to the atmosphere_xIs discharged
You will be able to prevent it. Also, NO_xAbsorbent
Since 18 has the function of a reduction catalyst,
NO even if the air-fuel ratio is the theoretical air-fuel ratio_xRelease from absorbent 18
NO done_xIs reduced. However, inflow and outflow
NO when the air-fuel ratio of air gas is set to the theoretical air-fuel ratio_xabsorption
Agent 18 to NO_xNO is released only gradually
_xAll NO absorbed in the absorbent 18_xTo release
Takes a little longer.

As described above, in the embodiment according to the present invention, the air-fuel ratio of the air-fuel mixture is basically controlled according to the value of the correction coefficient K distinguished by the boundaries Q, R and S in FIG. Therefore, in the lean combustion region (K <1.0) between the boundary line Q and the boundary line R, NO _x is absorbed by the NO _x absorbent 18, and the theoretical air-fuel ratio or rich combustion on the higher load side than the boundary line R is increased. area (K = 1.0 or K> 1.0) in the release _the NO _x absorbent 18 from the NO _x, than the boundary line Q even the low load side the stoichiometric air-fuel ratio combustion region (K = 1.0) NO _x absorbent 1
NO _x is released from 8.

[0026] releasing action of the NO _x from _the NO _x absorbent 18 during deceleration operation in order to increase the opportunities for NO _x emissions from _the NO _x absorbent 18 in this embodiment of the present invention is performed. That is, as shown in FIG. 6, when the throttle valve 14 is closed, the deceleration operation is started and the fuel supply is stopped (fuel cut), the fuel supply is restarted (return processing). The air-fuel ratio of the air-fuel mixture is temporarily rich (K
_t > 1.0), and at this time the NO _x absorbent 18 to N
O _x is released.

FIG. 7 shows a change in the correction coefficient K shown in FIG. 3 under a certain constant rotation speed N. As can be seen from FIG. 7, the air-fuel ratio of the air-fuel mixture is considerably lean in the lean combustion region between the boundary line Q and the boundary line R, and particularly, in the boundary line R, the air-fuel ratio is low on the low load side and the high load side. It can be seen that there is a considerable step in the magnitude of the fuel ratio. Therefore, when shifting from the lean combustion region to the stoichiometric air-fuel ratio combustion region beyond the boundary line R, a shock occurs because the output torque of the engine rapidly increases, and similarly, from the stoichiometric air-fuel ratio combustion region to the lean line region R. When shifting to the combustion region, a shock occurs because the output torque of the engine sharply decreases. Therefore, in the embodiment according to the present invention, the air-fuel ratio is changed from the lean combustion region beyond the boundary line R to the stoichiometric air-fuel ratio combustion region and from the stoichiometric air-fuel ratio combustion region beyond the boundary line R to the lean combustion region. I am trying to change it gradually.

However, if there is a step in the air-fuel ratio in this way,
As described with reference to FIG. 19, the combustion of lean mixture and
Combustion of air-fuel mixture with stoichiometric air-fuel ratio is repeated alternately.
As mentioned above, if the air-fuel ratio is gradually changed, it will often occur.
Amount NO_xWill be released into the atmosphere. That is, before
NO as mentioned_xNO_xTo absorb into the absorbent 18
Requires oxygen, but as shown in FIG.
As the degree decreases, the oxygen concentration contained in the exhaust gas
descend. Therefore NO_xNO by absorbent 18 _xAbsorption rate
As shown by the solid line in Fig. 8, the degree of lean is reduced.
The lower it gets, the lower it gets. On the other hand, NO emitted from the engine_xQuantity is
As shown by the broken line in FIG. 8, the air-fuel ratio is slightly higher than the theoretical air-fuel ratio.
Highest when lean. Therefore, the air-fuel ratio is the theoretical air-fuel
NO when slightly leaner than the ratio_xN by the absorbent 18
O_xCan no longer be absorbed, and thus NO_xIs the atmosphere
Will be discharged inside.

Therefore, the air-fuel ratio is gradually increased when shifting from the lean combustion region to the stoichiometric air-fuel ratio combustion region over the boundary line R and when shifting from the stoichiometric air-fuel ratio combustion region to the lean combustion region over the boundary line R. alters the air-fuel ratio will pass through slightly lean region than the stoichiometric air-fuel ratio, NO _x at this time is to be discharged into the atmosphere and thus. Therefore, in order to reduce the amount of NO _x emitted into the atmosphere, it is only necessary to reduce the chance that the air-fuel ratio becomes slightly leaner than the stoichiometric air-fuel ratio.

Therefore, in the first embodiment according to the present invention, FIG.
As shown in (A) and FIG. 10 (A), the boundary line R is composed of a first boundary line R and a second boundary line R'on the lower load side than the first boundary line R. , When the lean mixture is burned, the air-fuel ratio is controlled based on the correction coefficient K divided by the solid lines Q, R, and S in FIG. Lean combustion region (K <
1.0) beyond the boundary line R to the theoretical air-fuel ratio combustion region (K
= 1.0), the solid line Q in FIG.
The air-fuel ratio is controlled on the basis of the correction coefficient K classified by R'and S, and the theoretical air-fuel ratio combustion region (K
= 1.0) and beyond the boundary line R ', the lean combustion region (K
When shifting to <1.0), the air-fuel ratio is controlled again based on the correction coefficient K divided by the solid lines Q, R, and S in FIG. 9 (A).

The solid lines S, R and Q in FIG.
The correction coefficient K divided by is previously stored in the ROM 32 in the form of a map I shown in FIG.
In FIG. 10A, the correction coefficient K divided by the solid lines S, R ′, and Q is RO in advance in the form of map II shown in FIG. 10B.
It is stored in M32. As described above, when the map I and the map II are used properly, the air-fuel ratio does not repeat between the lean and stoichiometric air-fuel ratios, and thus the amount of NO _x released into the atmosphere can be greatly reduced. That is, as shown in FIG. 11, when the lean air-fuel mixture is being burned based on the map I, the throttle opening degree is slightly increased to slightly increase the vehicle speed, and at this time, the operating region of the engine is demarcated. correction coefficient K and the transitions to the stoichiometric air-fuel ratio combustion region beyond the line R becomes to be calculated based on the map II, final correction coefficient K _t to determine the actual fuel injection time is gradually Be increased to.

The throttle opening for the air-fuel ratio speed than the driver increases in the engine output Ru switching <br/> changeover Wa from the lean air-fuel ratio to the stoichiometric air-fuel ratio is greatly were expecting to increase Can be reduced. However, at this time, the correction coefficient K is the map
Since it is calculated based on II, the air-fuel ratio is maintained at the stoichiometric air-fuel ratio. Then the small throttle opening in order to decelerate, the final correction coefficient K _t shifting to the operating region lean combustion region beyond the boundary line R 'of the engine is to gradually decrease, the correction coefficient K is again mapped I It is calculated based on.

FIGS. 12 and 13 show an air-fuel ratio control routine for executing the first embodiment, and this routine is executed by interruption at fixed time intervals, for example. Referring to FIG. 12 and FIG. 13, first, step 10 is performed.
0 from the map shown in FIG. 2 to the basic fuel injection time TP
Is calculated. Next, at step 101, it is judged if the fuel supply is stopped during the deceleration operation, that is, if the fuel is being cut. When the fuel cut is not in progress, the routine proceeds to step 102, where it is judged whether or not the process for restarting the fuel supply, that is, the return process. If not in the process of returning, go to step 103 and map
It is determined whether or not the selection flag indicating that II should be selected is set. When the routine proceeds to step 103 for the first time after the engine is started, the selection flag has been reset, so the routine proceeds to step 104.

In step 104, the correction coefficient K is calculated from the map I shown in FIG. 9 (B). Then step 10
At 6, it is determined whether the correction coefficient K has changed from K <1.0 to K ≧ 1.0. If it has not changed, the routine proceeds to step 107, where the correction coefficient K is from K ≧ 1.0 to K <1.
It is determined whether or not it has changed to 0. When it has not changed, the routine proceeds to step 119, where it is judged if the control flag is set or not. At this time, since the control flag is not set, the routine jumps to step 108 and the correction coefficient K is set as the final correction coefficient K _t . Next, at step 109, it is judged if the final correction coefficient K _t is 1.0. When K _t <1.0 or K _t > 1.0, the routine proceeds to step 110, where the feedback correction coefficient FA
F is fixed to 1.0, and then in step 112, the fuel injection time TAU is calculated based on the following equation.

[0035] _{TAU = f · TP · K t} · FAF air this time is a lean air-fuel ratio or a rich air-fuel ratio according to the correction coefficient K. On the other hand, in step 109, K
_When it is judged that _t = 1.0, that is, when the stoichiometric air-fuel ratio mixture should be burned, the routine proceeds to step 111, where the feedback correction coefficient FAF is calculated based on the output signal of the air-fuel ratio sensor 21, and then step 112. Proceed to. In step 111, FAF is decreased when the air-fuel ratio sensor 21 detects that the air-fuel ratio has become rich, and FAF is detected when it is detected that the air-fuel ratio has become lean.
Is increased, so that the air-fuel ratio is maintained at the stoichiometric air-fuel ratio. Next, at step 113, the correction coefficient K is made K ₀ .

On the other hand, in step 106, the correction coefficient K
Is determined to have changed from K <1.0 to K ≧ 1.0, that is, when it is determined that the operating region of the engine has changed from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region, the routine proceeds to step 114. The selection flag is set with. Next, at step 116, the difference between the correction coefficient K ₀ in the previous processing cycle and the correction coefficient K calculated this time (K ₀ −
The value obtained by dividing K) by a constant integer n is ΔK (= (K ₀ −
K) / n). At this time, K ₀ <K, so ΔK
The value of is negative. Next, at step 117, K ₀ to Δ
The final correction coefficient K _t is calculated by subtracting K. Next, the routine proceeds to step 118, where the control flag is set, and then the routine proceeds to step 109.

In the next processing cycle, the routine proceeds from step 103 to step 105 where the correction coefficient K is calculated from the map II shown in FIG. 10 (B). Then steps 106, 1
After 07, the process proceeds to step 119. At this time, the control flag is set, so the routine proceeds to step 120, where K _t
It is determined whether or not the absolute value | K _t −K | of the difference between K and K has become smaller than | ΔK |. _{| K t -K |> | ΔK} |
In the case of, the routine proceeds to step 121, where ΔK is subtracted from K _t . At this time, since ΔK is negative, the value of K _t is increased by ΔK. Then, it proceeds to step 109.

Next, at step 120, | K _t −K
If it is determined that | ≦ | ΔK |, step 1
The control flag is reset at 22. Next, in step 108, the correction coefficient K calculated from the map II.
Is designated as K _t . Therefore, the correction coefficient K is K <1.0 to K
A map for calculating the correction coefficient K when the engine operating region changes from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region when the change is ≧ 1.0, that is, the map I
To Map II, and the final correction coefficient K _t is K
From (<1.0) to K (≧ 1.0), it is gradually increased by | ΔK |. Thus, the air-fuel ratio of the air-fuel mixture is gradually reduced.

Next, at step 107, the correction coefficient K
Is determined to have changed from K ≧ 1.0 to K <1.0, that is, it is determined that the operating region of the engine has changed from the stoichiometric air-fuel ratio or the rich combustion region to the lean combustion region, the routine proceeds to step 115. The selection flag is reset with. Next, at step 116, the difference (K ₀ −K) between the correction coefficient K ₀ in the previous processing cycle and the correction coefficient K calculated this time.
Is divided by a constant integer n to obtain ΔK (= (K ₀ −K) /
n). At this time, since K ₀ > K, the value of ΔK becomes positive. Next, at step 117, the final correction coefficient K _t is calculated by subtracting ΔK from K ₀ . Next, the routine proceeds to step 118, where the control flag is set, and then the routine proceeds to step 109.

In the next processing cycle, the routine proceeds from step 103 to step 104, and the correction coefficient K is calculated from the map I shown in FIG. 9 (B). Then steps 106 and 10
After 7, the process proceeds to step 119. Since the control flag has been set at this time, the routine proceeds to step 120, where it is judged if the absolute value | K _t −K | of the difference between K _t and K has become smaller than | ΔK |. _{| K t -K |> | ΔK} | subtracting [Delta] K from K _t proceeds to step 121 when the. At this time, since ΔK is positive, the value of K _t is decreased by ΔK. Then, it proceeds to step 109.

Next, at step 120, | K _t −K
If it is determined that | ≦ | ΔK |, step 1
The control flag is reset at 22. Next, in step 108, the correction coefficient K calculated from the map I is calculated.
Is designated as K _t . Therefore, the correction coefficient K is from K ≧ 1.0 to K
When the engine operating region changes to <1.0, that is, when the engine operating region shifts from the stoichiometric air-fuel ratio or the rich combustion region to the lean combustion region, the map for calculating the correction coefficient K is Map II.
To map I, and the final correction coefficient K _t is K
From (≧ 1.0) to K (<1.0), it is gradually decreased by | ΔK |. Thus, the air-fuel ratio of the air-fuel mixture is gradually increased.

On the other hand, when it is judged in step 101 that the fuel supply is in the deceleration operation state in which the fuel supply should be stopped, the processing cycle is completed, and thus the fuel injection is stopped. Next, when it is determined in step 102 that the operating state in which the fuel supply should be restarted has been reached, step 1
23, the recovery process is performed, and then step 112
Proceed to. As shown in FIG. 6, this return process is a process for temporarily setting the value of the final correction coefficient K _{t to} a value larger than 1.0. Therefore, when the fuel supply is restarted, the mixture gas The air-fuel ratio is temporarily made rich. When the return process is completed, the idling operation state is entered, so that the air-fuel ratio of the air-fuel mixture becomes the theoretical air-fuel ratio.

Next, a second embodiment according to the present invention will be described with reference to FIGS. 14 to 18. In the second embodiment, the correction coefficient K shown in FIG. 14 (A) is always used, and the correction coefficient K is stored in advance in the ROM 32 in the form of the map shown in FIG. 14 (B). As described above, in order to prevent the release of a large amount of NO _x into the atmosphere, it is necessary to prevent the combustion of the lean air-fuel mixture and the combustion of the air-fuel mixture with the stoichiometric air-fuel ratio from being alternately repeated. There is. For that purpose, when there is a possibility that the engine operating region alternates between the lean combustion region and the stoichiometric air-fuel ratio or the rich combustion region, the air-fuel ratio of the air-fuel mixture is changed to the lean air-fuel ratio or the theoretical air-fuel ratio regardless of the engine operating region. Alternatively, it may be temporarily fixed to the rich air-fuel ratio, that is, if the air-fuel ratio of the air-fuel mixture is fixed, the air-fuel ratio of the air-fuel mixture does not pass through a region slightly leaner than the stoichiometric air-fuel ratio, so NO _This will prevent _x from being emitted into the atmosphere.

Therefore, in this second embodiment, after the operating region of the engine has changed from the lean combustion region to the stoichiometric air-fuel ratio or rich combustion region, the operating region of the engine alternates between the lean combustion region and the stoichiometric air-fuel ratio or rich combustion region. If it is estimated that there is a possibility that the operating region of the engine will repeatedly alternate between the lean combustion region and the stoichiometric air-fuel ratio or the rich combustion region, the air-fuel ratio of the air-fuel mixture will be temporarily changed. I am trying to fix it.

By the way, various methods are conceivable for estimating that the operating region of the engine may alternately repeat the lean combustion region and the stoichiometric air-fuel ratio or rich combustion region. However, an example will be described. That is, as shown in FIG. 15, when the lean air-fuel mixture is being combusted, the throttle opening is slightly increased to slightly increase the vehicle speed, and at this time, the operating region of the engine exceeds the boundary line R. If it shifts to the stoichiometric air-fuel ratio combustion region, the engine output greatly increases at this time, and the vehicle speed increases more than the driver expected. Therefore, since the throttle opening is reduced, the operating region of the engine again crosses the boundary line R and shifts to the lean combustion region. In such a case, the amount of change in the engine load is small between the time when the operating region of the engine shifts to the stoichiometric air-fuel ratio combustion region and the time when it shifts to the lean combustion region again. It can be estimated from time to time that the operating region of the engine may alternately repeat the lean combustion region and the stoichiometric air-fuel ratio or rich combustion region.

When there is a possibility that the lean operating region and the stoichiometric air-fuel ratio or rich burning region are alternately repeated in the operating region of the engine, it will take a short time after the operating region of the engine shifts to the stoichiometric air-fuel ratio burning region. In the meantime, it shifts to the lean combustion region again. Therefore, in the second embodiment of the present invention, the period from when the operating region of the engine shifts to the stoichiometric air-fuel ratio burning region to when it shifts to the lean burning region again is shorter than a predetermined period C _t , and during this period. It is estimated that the operating region of the engine may alternately repeat the lean combustion region and the stoichiometric air-fuel ratio or rich combustion region when the variation amount of the engine load is small.

Further, in this embodiment, the variation amount of the engine load is calculated from the change of the throttle opening, and when the operating region of the engine is changed from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region. Throttle opening TA
_{As 0} , the maximum variation amount ΔTA based on the throttle opening TA ₀ is used as the variation amount of the engine load. When the maximum fluctuation amount ΔTA exceeds a predetermined amount, as shown in FIG. 15, during a predetermined constant period C ₀ ,
The air-fuel ratio of the air-fuel mixture is maintained at the stoichiometric air-fuel ratio ( _Kt = 1.0).

FIGS. 16 to 18 show an air-fuel ratio control routine for executing the second embodiment. This routine is executed by interruption at regular time intervals, for example. Figure 1
6 to 18, first, at step 200, the basic fuel injection time TP is calculated from the map shown in FIG. Next, at step 201, it is judged if the fuel supply is stopped during the deceleration operation, that is, if the fuel is being cut. When the fuel cut is not in progress, the routine proceeds to step 202, where it is judged whether or not the process for restarting the fuel supply, that is, the return process. When the restoration process is not in progress, the routine proceeds to step 203 and FIG.
The correction coefficient K is calculated from the map shown in (B), and then the count value C is incremented by 1 in step 204. Next, at step 205, it is judged if the stoichiometric flag indicating that the air-fuel ratio of the air-fuel mixture should be the stoichiometric air-fuel ratio is set or not. Since the normal stoichiometric flag has been reset, the routine proceeds to step 206.

In step 206, the correction coefficient K is K <1.
It is determined whether 0 has changed to K ≧ 1.0. If it has not changed, the routine proceeds to step 207, where the correction coefficient K
It is determined whether or not has changed from K ≧ 1.0 to K <1.0. If not changed, steps 208, 209
Then, the maximum variation amount ΔTA _max of the engine load is calculated, but at this time, the maximum variation amount ΔTA _max is merely calculated and is not used at all.

Next, the routine proceeds to step 210, where it is judged if the control flag is set or not. At this time, since the control flag is not set, the routine jumps to step 211 and the correction coefficient K is set as the final correction coefficient K _t .
Next, at step 212, the final correction coefficient K _t is 1.
It is determined whether it is 0 or not. _Kt <1.0 or _Kt >
When it is 1.0, the routine proceeds to step 213, where the feedback correction coefficient FAF is fixed to 1.0, and then, at step 215, the fuel injection time TAU is calculated based on the following equation.

[0051] _{TAU = f · TP · K t} · FAF air this time is a lean air-fuel ratio or a rich air-fuel ratio according to the correction coefficient K. On the other hand, in step 212, K
_When it is determined that _t = 1.0, that is, when the stoichiometric air-fuel ratio mixture should be burned, the routine proceeds to step 214, where the feedback correction coefficient FAF is calculated based on the output signal of the air-fuel ratio sensor 21, and then step 215. Proceed to. In step 214, FAF is decreased when the air-fuel ratio sensor 21 detects that the air-fuel ratio has become rich, and FAF is detected when it is detected that the air-fuel ratio has become lean.
Is increased, so that the air-fuel ratio is maintained at the stoichiometric air-fuel ratio. Next, at step 216, the correction coefficient K _t
Is designated as K ₀ .

On the other hand, in step 206, the correction coefficient K
Is determined to have changed from K <1.0 to K ≧ 1.0, that is, when it is determined that the operating region of the engine has changed from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region, the routine proceeds to step 217. Then, the throttle opening TA detected by the throttle opening sensor 20 is set to TA ₀ (FIG. 15). Next, at step 218, the count value C and the maximum fluctuation amount ΔTA _{max of the} engine load are made zero.

Next, at step 219, the correction coefficient K ₀ in the previous processing cycle and the correction coefficient K calculated this time are calculated.
The value obtained by dividing the difference (K ₀ −K) with a constant integer n is ΔK.
(= (K ₀ −K) / n). At this time, since K ₀ <K, the value of ΔK becomes negative. Next, at step 220, the final correction coefficient K _t is calculated by subtracting ΔK from K ₀ . Next, the routine proceeds to step 221, the control flag is set, and then the routine proceeds to step 212.

In the next processing cycle, steps 206 and 2
07, the routine proceeds to step 208, where TA ₀ (FIG. 15) is subtracted from the current throttle opening TA (TA-T
It is determined whether A ₀ ) is larger than the maximum fluctuation amount ΔTA _max . When TA−TA ₀ > ΔTA _max , the routine proceeds to step 209, where ΔTA _max = TA−TA ₀ .
Therefore, it can be seen that ΔTA _max represents the maximum variation of the throttle opening after TA = TA ₀ , that is, the maximum variation of the engine load.

Next, the routine proceeds to step 210, at which time the control flag has been set, so the routine proceeds to step 222, where it is determined whether the absolute value | K _t -K | of the difference between K _t and K has become smaller than | ΔK |. Is determined. | K _t -K |> |
When ΔK |, the routine proceeds to step 223, where K _{t is changed} to ΔK.
Subtract. At this time, since ΔK is negative, the value of K _t is Δ
It can be increased by K. Then, it proceeds to step 212.

Next, at step 222, | K _t −K
If it is determined that | ≦ | ΔK |, step 2
The control flag is reset at 24. Then the correction coefficient K is calculated from the map shown in FIG. 14 (B) is a K _t proceeds to step 224. Therefore, when the correction coefficient K changes from K <1.0 to K ≧ 1.0, that is, when the operating region of the engine shifts from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region, the final correction coefficient K _t is From K (<1.0) to K (≧ 1.0) is gradually increased by | ΔK |, and thus the air-fuel ratio of the air-fuel mixture is gradually reduced.

Next, at step 207, the correction coefficient K
Is determined to have changed from K ≥ 1.0 to K <1.0, that is, when it is determined that the operating region of the engine has shifted from the stoichiometric air-fuel ratio or the rich combustion region to the lean combustion region, the routine proceeds to step 225. At, it is determined whether or not the count value C is smaller than a predetermined constant value C _t . When C ≦ C _t, the routine proceeds to step 226, where it is judged if the maximum fluctuation amount ΔTA _max is smaller than a predetermined constant amount α. If C> C _t or ΔTA _max ≧ α, the process proceeds to step 219.

In step 219, the previous processing cycle
Correction coefficient K₀And the difference between the correction coefficient K calculated this time
(K₀The value obtained by dividing −K) by a constant integer n is ΔK (=
(K₀-K) / n). At this time K₀> K
Therefore, the value of ΔK becomes positive. Next, at step 220, K
₀The final correction factor K by subtracting ΔK from
_tIs calculated. Then, control proceeds to step 221.
The lag is set, then proceed to step 212.

In the next processing cycle, steps 206 and 2
07, and then steps 208 and 209.
Go to step 210. At this time the control flag is set
Yes, go to step 222 and K_tThe difference between K and K
Log value | K_tWhether -K | has become smaller than | ΔK |
Is determined. ｜ K_t-K |> | ΔK |
Go to 223 K_tIs subtracted from ΔK. At this time
ΔK is positive, so K _tThe value of is reduced by ΔK
It Then, it proceeds to step 212.

Next, at step 222, | K _t −K
If it is determined that | ≦ | ΔK |, step 2
The control flag is reset at 24. Then the correction coefficient K is calculated from the map shown in FIG. 14 (B) is a K _t proceeds to step 211. Therefore, when the correction coefficient K changes from K ≧ 1.0 to K <1.0, that is, when the operating region of the engine shifts from the stoichiometric air-fuel ratio or the rich combustion region to the lean combustion region, C> C _t or ΔTA _max. If ≧ α, the final correction coefficient K _t is K (≧ 1.0) to K
The air-fuel ratio of the air-fuel mixture is gradually increased by gradually decreasing by | ΔK | until (<1.0).

On the other hand, when the operating region of the engine shifts from the stoichiometric air-fuel ratio or the rich combustion region to the lean combustion region and C ≦ C _t and ΔTA _max <α, that is, the operating region of the engine is lean. When it is estimated that the combustion region and the stoichiometric air-fuel ratio or the rich combustion region may be alternately repeated, the routine proceeds to step 227, where the final correction coefficient K _t is fixed at 1.0, and then the routine proceeds to step 228. The stoichiometric flag is set. Then, it proceeds to step 212.

In the next processing cycle, the routine proceeds from step 205 to step 229, where it is judged if the count value C has exceeded a predetermined constant time C ₀ . When C <C _0, the routine jumps to step 212, and when C ≧ C ₀ , the routine proceeds to step 230, where the stoichiometric flag is reset, and then the routine proceeds to step 212. Therefore, as shown in FIG. 15, the final correction coefficient K _t is maintained at 1.0 until a certain period of time C ₀ has elapsed after the operating region of the engine changed from the lean combustion region to the stoichiometric air-fuel ratio or the rich combustion region. Therefore, during this period C ₀ , the air-fuel mixture is maintained at the stoichiometric air-fuel ratio.

On the other hand, when it is judged in step 201 that the fuel supply is in the decelerating operation state in which the fuel supply should be stopped, the processing cycle is completed, and thus the fuel injection is stopped. Next, when it is determined in step 202 that the operating state in which the fuel supply should be restarted has been reached, step 2
Then, the process proceeds to step 31, and a return process is performed, and then step 215
Proceed to. As shown in FIG. 6, this return process is a process for temporarily setting the value of the final correction coefficient K _{t to} a value larger than 1.0. Therefore, when the fuel supply is restarted, the mixture gas The air-fuel ratio is temporarily made rich. When the return process is completed, the idling operation state is entered, so that the air-fuel ratio of the air-fuel mixture becomes the theoretical air-fuel ratio.

[0064]

The NO _x can be prevented from being discharged into the atmosphere without being absorbed by the NO _x absorbent.

[Brief description of drawings]

FIG. 1 is an overall view of an internal combustion engine.

FIG. 2 is a diagram showing a map of a basic fuel injection time.

FIG. 3 is a diagram showing a correction coefficient K.

FIG. 4 Unburned HC and C in exhaust gas discharged from the engine
It is a diagram which shows the concentration of O and oxygen roughly.

FIG. 5 is a diagram for explaining the action of absorbing and releasing NO _x .

FIG. 6 is a diagram for explaining a recovery process during deceleration operation.

FIG. 7 is a diagram showing a correction coefficient K.

FIG. 8 is a diagram showing a NO _x absorption rate and a NO _x emission amount.

FIG. 9 is a diagram showing a correction coefficient K.

FIG. 10 is a diagram showing a correction coefficient.

FIG. 11 is a time chart of air-fuel ratio control.

FIG. 12 is a flowchart showing air-fuel ratio control.

FIG. 13 is a flowchart showing air-fuel ratio control.

FIG. 14 is a diagram showing a correction coefficient K.

FIG. 15 is a time chart of air-fuel ratio control.

FIG. 16 is a flowchart showing air-fuel ratio control.

FIG. 17 is a flowchart showing air-fuel ratio control.

FIG. 18 is a flowchart showing air-fuel ratio control.

FIG. 19 is a time chart for explaining conventional problems.

[Explanation of symbols]

16 ... Exhaust pipe 18 ... NO _x absorbent

Front Page Continuation (72) Inventor Satoshi Iguchi 1 Toyota Town, Toyota City, Aichi Prefecture Toyota Motor Corporation (56) References JP 57-7741 (JP, A) JP 58-214648 (JP, A) International publication 93/020806 (WO, A1) (58) Fields investigated (Int.Cl. ⁷ , DB name) F02D 41/04 305 F01N 3/08 F02D 41/14 310

Claims (3)

(57) [Claims] 1. A absorbs NO _x when the air-fuel ratio of the inflowing exhaust gas is lean, the engine and _the NO _x absorbent air-fuel ratio of the exhaust gas to release the NO _x absorbed when the stoichiometric air-fuel ratio or rich flowing Located in the exhaust passage, the operating region of the engine is divided into a lean combustion region on the low load side and a stoichiometric air-fuel ratio or rich combustion region on the high load side.In the lean combustion region, the lean air-fuel mixture is burned to remove the engine from the engine. allowed absorb the discharged NO _x into _the NO _x absorbent, when the stoichiometric air-fuel ratio or rich combustion region is absorbed in _the NO _x absorbent by burned air-fuel mixture or a rich mixture of the theoretical air-fuel ratio N
In an internal combustion engine which is adapted to release O _x, with gradually decreasing the air-fuel ratio of the mixture when the operating region of the engine changes from the lean combustion region to the stoichiometric air-fuel ratio combustion region
The engine operating range is from the stoichiometric air-fuel ratio combustion range to the lean combustion range.
Comprising a gradually increasing to <br/> air gradual change means the air-fuel ratio of the mixture when going to pass, a lean combustion region and the stoichiometric air-fuel ratio
The boundary line between the combustion region and the lean combustion region is the first boundary line at the transition to the stoichiometric air-fuel ratio combustion region, and the second boundary line is the boundary at the transition from the stoichiometric air-fuel ratio combustion region to the lean combustion region. An exhaust emission control device for an internal combustion engine, which is configured by a boundary line and has a second boundary line set on a low load side with respect to the first boundary line. Wherein absorbing the NO _x when the air-fuel ratio of the inflowing exhaust gas is lean, the engine and _the NO _x absorbent air-fuel ratio of the exhaust gas to release the NO _x absorbed when the stoichiometric air-fuel ratio or rich flowing Located in the exhaust passage, the operating region of the engine is divided into a lean combustion region on the low load side and a stoichiometric air-fuel ratio or rich combustion region on the high load side.In the lean combustion region, the lean air-fuel mixture is burned to remove the engine from the engine. allowed absorb the discharged NO _x into _the NO _x absorbent, when the stoichiometric air-fuel ratio or rich combustion region is absorbed in _the NO _x absorbent by burned air-fuel mixture or a rich mixture of the theoretical air-fuel ratio N
In an internal combustion engine which is adapted to release O _x, with gradually decreasing the air-fuel ratio of the mixture when the operating region of the engine changes from the lean combustion region to the stoichiometric air-fuel ratio combustion region
The engine operating range is from the stoichiometric air-fuel ratio combustion range to the lean combustion range.
The air-fuel ratio of the air-fuel mixture is gradually increased when shifting to the region. <br /> Air-fuel ratio gradual changing means and the engine operating region after the operating region of the engine shifts from the lean combustion region to the stoichiometric air-fuel ratio combustion region. It is estimated that there is a possibility that the lean combustion region and the stoichiometric air-fuel ratio combustion region may be repeated alternately, and that the operating region of the engine may alternate between the lean combustion region and the stoichiometric air-fuel ratio combustion region. And an air-fuel ratio fixing means for temporarily fixing the air-fuel ratio of the air-fuel mixture to the lean air-fuel ratio or the stoichiometric air-fuel ratio . 3. The estimating means preliminarily determines an amount of change in engine load before the operating region of the engine shifts to the lean combustion region again after the operating region of the engine shifts from the lean combustion region to the stoichiometric air-fuel ratio combustion region. The exhaust emission control device for an internal combustion engine according to claim 2, wherein it is estimated that the operating region of the engine may repeat the lean combustion region and the stoichiometric air-fuel ratio combustion region alternately when the amount is smaller than a predetermined amount.