JPH0526080A - Air-fuel ratio control method for lean burn engine system - Google Patents

Abstract

PURPOSE:To prevent the lowering of NOx purifying ability when air-fuel ratio control is changed from a lean control mode to a theory air fuel ratio control mode, by controlling the air-fuel ratio on a a rich side until the output of a rear oxygen sensor is reversed to the theory air fuel ratio side from the lean side. CONSTITUTION:A catalyst converter 3 housing a lean NOx catalyst 4 and a ternary catalyst 5 is furnished at the middle of an exhaust gas pipe passage 2, and front and rear oxygen sensors 6, 7 are provided at the upstream and downstream parts of this converter 3. And an air fuel ratio is controlled by means of an ECU 8 on the basis of the output of the front sensor 6, and also the control of the air-fuel ratio is corrected on the basis of the output of the rear sensor 7, and at the same time the air fuel ratio is controlled by changing it to a lean control mode A or a theory air-fuel ratio control mode B according to an engine operation situation. On the occasion of this air-fuel ratio control, when the air-fuel ratio is changed to the B from the mode A, arrangement is made so that the air-fuel ratio may be controlled on the rich side until the output of the rear sensor 7, from the viewpoint of the air-fuel ratio, is changed to the theory air-fuel ratio side from the lean side.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention particularly relates to an air-fuel ratio control method for a lean burn engine system for maintaining the purification efficiency of a catalytic converter.

[0002]

2. Description of the Related Art A lean burn engine system controls the air-fuel ratio of the engine in a lean control mode when the engine is in a steady operation state, while the engine is in a transitional operation state such as during acceleration. The air-fuel ratio is to be controlled in the stoichiometric air-fuel ratio control mode. If the air-fuel ratio is controlled in this way, CO and HC in the exhaust gas
It is possible to reduce fuel consumption and improve fuel efficiency.

However, when the air-fuel ratio is controlled in the lean control mode, since oxygen is present in the exhaust gas, a catalytic converter containing a three-way catalyst or the like is inserted in the exhaust gas pipeline. However, this catalytic converter cannot efficiently reduce and purify NOx,
There is a problem that the amount of Ox emission increases. For this reason, in the exhaust gas pipeline of the lean burn engine system, which is located upstream of the three-way catalyst of the catalytic converter and reduces NOx in the exhaust gas to purify it even in the presence of oxygen. A NOx catalyst is arranged.

On the other hand, in order to make the three-way catalyst of the catalytic converter work efficiently, it is necessary to control the air-fuel ratio of the engine to the stoichiometric air-fuel ratio with high accuracy, and therefore, in the exhaust gas pipeline of the lean burn engine system described above. Has a front oxygen sensor and a rear oxygen sensor, respectively. The front oxygen sensor is located close to the engine side, whereas
The rear oxygen sensor is located downstream of the catalytic converter. Thus, if the lean burn engine system is provided with two oxygen sensors, the air-fuel ratio of the engine is controlled based on the oxygen concentration in the exhaust gas detected by the front oxygen sensor, while the exhaust gas detected by the rear oxygen sensor is controlled. The air-fuel ratio can be corrected and controlled on the basis of the oxygen concentration in the air, which makes it possible to control the air-fuel ratio of the engine to the stoichiometric air-fuel ratio with high accuracy.

[0005]

In the lean burn engine system described above, when the lean NOx catalyst arranged in the exhaust gas pipe has a large oxygen adsorption amount, the air-fuel ratio of the engine changes from the lean control mode to the theoretical air-fuel ratio. Immediately after switching to the fuel ratio control mode, the adsorbed oxygen is released from the lean NOx catalyst for a while, so that the three-way catalyst of the catalytic converter does not function effectively.

Further, in this case, even if the air-fuel ratio is actually controlled to the stoichiometric air-fuel ratio, since oxygen is detected in the exhaust gas by the rear oxygen sensor, the rear oxygen sensor is based on the sensor signal. When the air-fuel ratio of the engine is corrected, the correction may excessively correct the air-fuel ratio to the rich side, which may cause so-called rich spike.

The present invention has been made based on the above-mentioned circumstances, and the purpose thereof is to detect NO in exhaust gas.
An object of the present invention is to provide an air-fuel ratio control method for a lean burn engine system, which can not only efficiently purify x, but also can appropriately perform air-fuel ratio control.

[0008]

The present invention is directed to a catalytic converter arranged in an exhaust gas pipe extending from an engine, and a lean NOx arranged in the exhaust gas pipe upstream of a catalyst of the catalytic converter. A catalyst, a front oxygen sensor located in the exhaust gas pipe upstream of the lean NOx catalyst, for detecting the oxygen concentration in the exhaust gas to control the air-fuel ratio, and in the exhaust gas pipe than the lean NOx catalyst. A rear oxygen sensor, which is located downstream and is used to correct the air-fuel ratio control by detecting the oxygen concentration in the exhaust gas, and based on the sensor signals from the front and rear oxygen sensors, depending on the operating conditions of the engine. The air-fuel ratio of the engine
In a lean burn engine system comprising a lean side control mode controlled on the lean side and a control means for switching control to a stoichiometric air-fuel ratio control mode for controlling to a stoichiometric air-fuel ratio, the air-fuel ratio control method of the present invention is When the air-fuel ratio is switched from the lean control mode to the stoichiometric air-fuel ratio control mode, the engine is operated until the oxygen concentration in the exhaust gas detected by the rear oxygen sensor changes from the lean side to the stoichiometric air-fuel ratio side in terms of the air-fuel ratio. The air-fuel ratio of is controlled on the rich side.

[0009]

According to the above-described air-fuel ratio control method, the air-fuel ratio control of the engine is performed immediately after switching from the rich control mode to the stoichiometric air-fuel ratio control mode, and the air-fuel ratio determined based on the sensor signal of the rear oxygen sensor is not yet reached. When on the lean side, the air-fuel ratio is controlled on the rich side, so in this case, the amounts of CO and HC in the exhaust gas of the engine increase. However, since these increased CO and HC are oxidized by the oxygen released from the lean NOx catalyst and consume the oxygen, the lean NOx is consumed.
The adverse effect of oxygen released from the x-catalyst on the catalytic action of the catalytic converter can be greatly reduced, and as a result, NOx in the exhaust gas is effectively purified by the three-way catalyst of the catalytic converter.

In the above situation, the rear oxygen sensor detects the oxygen concentration in the exhaust gas after the oxygen released from the lean NOx catalyst has already been consumed.
In this case, the oxygen concentration in the exhaust gas detected by the rear oxygen sensor is almost zero or low. Therefore, the sensor signals of the front and rear oxygen sensors both indicate that the air-fuel ratio of the engine is controlled to the stoichiometric air-fuel ratio.

The control on the rich side of the air-fuel ratio immediately after the start of the above-mentioned stoichiometric air-fuel ratio control is performed until the air-fuel ratio determined based on the sensor signal of the rear oxygen sensor is reversed from the rich side to the stoichiometric air-fuel ratio side. Will continue.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT Referring to FIG. 1, a lean burn engine system is schematically illustrated. The engine system comprises an engine 1 from which an exhaust gas line 2 extends. In the middle of this exhaust gas pipeline 2,
A catalytic converter 3 is inserted, and in the catalytic converter 3, a zeolite-based lean NOx catalyst 4 and a three-way catalyst 5 are accommodated in tandem. Here, as is apparent from FIG. 1, the lean NOx catalyst 4 is arranged on the engine 1 side with respect to the three-way catalyst 5 in the flow direction of the exhaust gas, that is, on the upstream side of the three-way catalyst 5. Has been done.

The exhaust gas pipe 2 is provided with front and rear oxygen sensors 6 and 7 for detecting the oxygen concentration in the exhaust gas. The front oxygen sensor 6 is positioned on the engine 1 side, and Rear oxygen sensor 7
Is positioned downstream of the catalytic converter 3 in the flow direction of the exhaust gas. The front and rear oxygen sensors 6 and 7 are used for feedback control of the air-fuel ratio of the engine 1. Therefore, the front and rear oxygen sensors 6 and 7 are provided in an electronic control unit (ECU) 8 as a control means. It is electrically connected and the sensor signal can be supplied to the ECU 8.

On the other hand, an intake passage 9 is connected to the engine 1, and a fuel injector 10 for ejecting fuel toward a combustion chamber (not shown) of the engine 1 is provided in the intake passage 9. .. This fuel injector 10 is also electrically connected to the ECU 8, and its operation, that is, the fuel injection amount is based on the sensor signals from the front and rear oxygen sensors 6 and 7, and the ECU 8
Is controlled by a control signal output from the fuel injector 10 to the fuel injector 10, and thus the air-fuel ratio of the engine 1 is feedback-controlled so as to be maintained at a desired value.

Further, the ECU 8 has a function of switching the air-fuel ratio control of the engine 1 between the lean control mode and the stoichiometric air-fuel ratio control mode. Therefore, the ECU 8 has the above-mentioned front and rear oxygen sensors 6, 6. In addition to 7, various sensors such as an accelerator opening sensor 11 and a brake pedal sensor 12 used to determine whether the engine 1 is in a steady operation state are also connected.

Next, the air-fuel ratio control routine executed by the ECU 8 will be described with reference to the flowcharts of FIGS. Air-fuel ratio control routine First, step S in FIG.
At 0, it is determined whether or not the engine 1 has been started, and the determination here is performed based on, for example, whether or not an ignition key (not shown) of the engine 1 is turned on. When the engine 1 is started and the determination result of step S0 becomes positive (YES), the initial control of the air-fuel ratio (A / F) is performed in the next step S1.

Step S1 comprises the subroutine shown in FIG. 3. When this subroutine is executed, it is just after the engine 1 is started. In this case, therefore, in step S100, the A / F is feedback-controlled to the stoichiometric air-fuel ratio using the sensor signal of only the front oxygen sensor 6. That is, step S1
At 00, the A / F is controlled in the stoichiometric air-fuel ratio control mode (S-FB) using only the front oxygen sensor 6 that is close to the engine 1 and is immediately activated.

Step S100 is continuously executed for a predetermined time. After that, when the rear oxygen sensor 7 is activated and the cooling water temperature of the engine 1 reaches the predetermined temperature, the next step S101 is executed. It In this step, the A / F will be controlled in S-FB mode using the sensor signals of the front and rear oxygen sensors 6,7. Therefore, in such a situation, the A / F
S-FB using only the sensor signal of the front oxygen sensor 6
Since the gain correction using the sensor signal of the rear oxygen sensor 7 is added to the mode for control, the A / F is controlled to the stoichiometric air-fuel ratio with high accuracy.

After the above-mentioned subroutine is executed, the routine proceeds to step S2, where it is judged if the engine 1 is in the driving state. If the determination result here is positive, the flag F1
It is determined whether or not 1 is set, but the determination result is negative at this point, and the process proceeds to step S4.
If the determination result of steps S0 and S2 is negative,
This air-fuel ratio control routine ends.

In step S4, the operating state of the engine 1 becomes a steady operating state, and the lean control mode (L-FB
It is determined whether or not the condition that the A / F feedback control in the mode) can be performed is satisfied. The determination here can be performed based on, for example, a sensor signal from the accelerator opening sensor 11 or the like. If the determination result of step S4 is positive, next, it is determined whether the S-FB mode is switched to the L-FB mode and the A / F is controlled. Normally, since the determination result of the previous step S4 is positive, the determination result is positive, and therefore the counting of the first timer is executed in the next step S6. The first timer here is step S5.
It measures the elapsed time since the positive.

After that, in step S7, it is judged whether or not the value of the first timer has reached the predetermined time t1 (≧ 0), but immediately after the judgment result of step S5 becomes positive, the judgment is made. Since the result is negative, in this case, step S
Go to 8. In step S8, contrary to the determination in step S5, it is determined whether the L-FB mode is switched to the S-FB mode and the A / F is controlled. In this situation, the determination is made. The result is negative, the process returns to step S3 and the above steps are repeated.

If the result of the determination in step S7 is negative while the results of the determination in steps S4 and S5 are maintained positive, the use of the rear oxygen sensor 7 is stopped in the next step S9. That is, at this point, the sensor signal of the rear oxygen sensor 7 is not used for the control of the A / F in the L-FB mode, and therefore the A / F control in the L-FB mode is
It will be implemented based on the sensor signal of the front oxygen sensor 6.

After the determination result of step S5 becomes positive, step S9 may be executed without executing steps S6 and S7. However, even if the A / F of the engine 1 becomes lean, there is some delay in determining the lean state based on the sensor signal of the rear oxygen sensor 7. Therefore, the amount of the delay, that is, the above If the use of the rear oxygen sensor 7 is stopped after confirming that the A / F shifts to the lean side by the rear oxygen sensor 7 for a predetermined time t1, the front oxygen sensor 6 should fail. Even then, the failure can be detected.

From step S9, the above-mentioned step S
Since the process proceeds to step 8, the A / F is controlled in the L-FB mode as long as the determination result of step S8 is maintained to be negative. However, the operating state of the engine 1 is in a transitional period from the steady operating state, such as during acceleration, and therefore the A / F is controlled by switching from the L-FB mode to the S-FB mode. In such a case, in this case, the determination result of step S8 is negative, the process proceeds from step S8 to step S10 of FIG. 4, the above-mentioned flag F1 is set to 1, and the above-mentioned first step is performed. 1 Timer value is cleared.

In the next step S11, the second timer is activated to start counting. That is, the second timer measures the elapsed time after the above-mentioned step S8 becomes positive. In the next step S12, it is determined whether or not the value of the second timer has reached the predetermined time t2.
Immediately after 8 becomes positive, the determination result is negative, and therefore, the process returns to step 2 in FIG. 2 and the steps after this step are repeated.

When steps S2 to S3 are executed, the flag F1 has already been set to 1, so the result of the determination is positive. Therefore, from step S3, the process jumps to step S8. The steps after this step are performed. After that, when the determination result of step S12 becomes positive, the process proceeds to the next step S13, and in this step, the use of the rear oxygen sensor 7 is restarted.

At the next step S14, it is judged if the A / F judged based on the sensor signal of the rear oxygen sensor 7 is reversed from the lean side to the stoichiometric air-fuel ratio side. Here, after the determination result of step S8 becomes positive, after execution of step S13 after the elapse of a predetermined time t2, step S13 is performed.
The reason for performing the determination of 14 is as follows. That is, the determination result of step S8 becomes positive, and even if the A / F of the engine 1 is controlled in the S-FB mode, there is a delay until the exhaust gas from the engine 1 reaches the rear oxygen sensor 7. The start of control in the mode cannot be immediately detected from the sensor signal of the rear oxygen sensor 7. Therefore,
If the determination in step S14 is performed after the elapse of the predetermined time t2, when the oxygen concentration is detected by the rear oxygen sensor 7, the oxygen is the oxygen once adsorbed by the lean NOx catalyst 4 described above. It can be determined that it is released from the catalyst.

Therefore, if the result of the determination in step S14 is positive, the process proceeds to step S15. In this step, the front and rear oxygen sensors 6 and 7 are used to perform normal S.
The A / F is feedback-controlled in the -FB mode. That is, in step S15, the same A / F control as in step S101 of FIG. 3 described above is performed, and in this case, the S-FB control is based on the sensor signal of the rear oxygen sensor 7 and uses a normal gain. Will be corrected. Note that this gain is based on the fuel injector 10
It is used to determine the amount of fuel injected from the.

After this, in the next step S16, the flag F1 is reset to 0, and at the same time, the values of the second and third timers are also cleared. The third timer will be described later. From step S16, the process returns to step S2 of FIG. 2 and the steps after this step are repeatedly executed, but at this time, the flag F1 is 0.
Since it is reset to, the determination result in the next step S3 is negative, and the process proceeds to step S4, and the determination in this step is performed.

In this case, when the operating state of the engine 1 is not in the steady operating state and is still in the transitional period, the determination result in step S4 is negative, and the process immediately proceeds from step S4 to step S15 in FIG. This gives A
The / F control will be executed while being maintained in the S-FB mode. On the other hand, when the determination result of step S14 is negative, that is, when the determination result of step S14 is negative after a predetermined time t2 has elapsed since the A / F control was switched from the L-FB mode to the S-FB mode. To step S
In step S17, the A / F is feedback-controlled in accordance with the S-FB mode by using the rich gain instead of the normal gain described above. It should be noted that the rich gain is determined in the step described later, and therefore, at this point, the S-FB mode on the rich side is not substantially executed.

In the next step S18, the learning control based on the sensor signal from the front oxygen sensor 6 is stopped. This learning control is necessary to compensate for variations in the output characteristics of the individual front oxygen sensors, but in the case of an air-fuel ratio control routine that does not include learning control, step S18 is omitted. be able to. However, if the front oxygen sensor has learning control, erroneous learning is performed unless step S18 is performed, and thus step S18 is necessary.

From step S18, step S of FIG.
Proceeding to 19, in this step, it is judged if the fuel injection of the engine 1 is asynchronous injection. This asynchronous injection is performed to increase the feeling of acceleration of the engine 1,
If the result of determination in step S19 is positive, step S19 of FIG.
Returning to 14, the steps after this step are repeatedly performed.

On the other hand, if the determination result in step S19 is negative, the process proceeds to step S20, and the rich gain P described above is determined in this step S20. In step S20, for example, the rich gain P is calculated based on the following equation. P (n + 1) = P (n) · (1-K) where P (n) is the current rich gain, P (n + 1) is the next rich gain, and n is a natural number of 0 or more. , K indicate tailing coefficients, respectively. The initial value P (0) of the rich gain P is set to P0, and the initial value of the tailing coefficient K is set to K1 (<1).

In the next step S21, as in the case of step S14 described above, it is judged whether or not the sensor signal of the rear oxygen sensor 7 is inverted from the lean side in A / F. In the case of, in step S22, after the third timer starts counting, in the next step S23, the value of the third timer is set to the predetermined time t3.
Is determined. Here, the third timer is
Substantially the above-described step S14 is performed, and the elapsed time from the case where the determination result is negative is measured.

If the determination result in step S23 is negative, the process returns to step S17 in FIG. 4 and the steps after this step are repeated. Therefore, in step S19,
As long as the determination result in S21 is maintained to be NO, the value of the rich gain P decreases with time as shown in FIG. 4 and is determined by the initial value K1 of the tailing coefficient K. Is reduced according to. Further, in this situation, by repeating the execution of step S17, the A / F is controlled in accordance with the S-FB mode based on the rich gain P, and as a result, the A / F once shifts to the rich side, and then theoretically. It will be controlled so as to approach the air-fuel ratio.

However, if the result of the determination in step S23 is positive, the tailing coefficient K is replaced with 0 in the next step S24, and then in step S25, the constant P1 (is added to the current rich gain P (n). <P0) is substituted.
Therefore, after the determination result of step S23 becomes positive, when the steps of step S17 and subsequent steps are repeated in the same manner as described above, the rich gain P becomes a constant value as shown in FIG. While being maintained, the A / F will be controlled in the rich side region.

On the other hand, if the result of the determination in step S21 is positive, the process proceeds to step S26, in which the values of the rich gain P and the tailing coefficient K are respectively replaced with the initial values, and then step S15 of FIG. Then, the steps after this step will be performed as described above. As described above, in the air-fuel ratio control routine of the embodiment, the A / F control is changed from the L-FB mode to S.
-When switching to the FB mode, the rich gain P is used for the A / F control in the S-FB mode while the sensor signal of the rear oxygen sensor does not reverse from the lean side to the stoichiometric air-fuel ratio side in A / F. Then, since the A / F is controlled substantially on the rich side, even if the oxygen adsorbed on the lean NOx catalyst is released, this oxygen causes an increase in the oxidation of CO and HC in the exhaust gas. It is used and consumed, and as a result, the oxygen concentration in the exhaust gas passing through the three-way catalyst 5 of the catalytic converter 3 can be reduced or eliminated, and the catalytic action of the three-way catalyst 5, particularly the NOx purification action, is a lean NOx catalyst. Exhaust gas can be effectively purified without being adversely affected by oxygen from 4.

In the above situation, the use of the dedicated rich gain P may undesirably excessively correct the S-FB mode based on the sensor signal from the rear oxygen sensor 7. Nor. Referring to FIG. 7, there is shown the NOx purification efficiency in the catalytic converter when the air-fuel ratio control routine of one embodiment is not executed,
The broken line A shows the NOx purification efficiency of the catalytic converter 3 having the lean NOx catalyst 4 and the three-way catalyst 5, while the solid line B shows the NOx purification efficiency of the catalytic converter having only the three-way catalyst. .. As is apparent from FIG. 7, the broken line A
The purification rate characteristic of No. 2 is lower in purification ability than the purification rate characteristic of solid line B, but it is considered that this is due to oxygen released from the lean NOx catalyst 4. However, if the air-fuel ratio control routine described above is executed, A /
After the F control is switched from the L-FB mode to the S-FB mode, when the sensor signal of the rear oxygen sensor 7 is on the lean side in terms of A / F, the lean side is used to control the A / F to lean the lean side. The adverse effect of oxygen released from the NOx catalyst 4 toward the three-way catalyst 5 can be removed, whereby the NOx purification efficiency of the three-way catalyst 5 can be sufficiently maintained. The hatched area in FIG. 7 indicates the so-called window width of the three-way catalyst 5, but as is clear from the above description, the air-fuel ratio control routine of this embodiment is executed. For example, it does not narrow the window width.

Further, the sensor signal of the rear oxygen sensor 7 is A
A / F on the theoretical air-fuel ratio side by reversing from the lean side in terms of / F
In the case of, the A / F is immediately controlled in the normal S-FB mode, so that the fuel efficiency is not deteriorated.
When the A / F is controlled in the L-FB mode,
Of course, NOx in the exhaust gas is effectively purified by the lean NOx catalyst, and since the use of the rear oxygen sensor 7 is stopped at this time, the sensor signal from the rear oxygen sensor 7 is L-FB. It does not adversely affect the control in the mode.

The present invention is not limited to the above-described embodiment, but various modifications can be made. For example, if the initial values of the rich gain P and the tailing coefficient K described above are learned and obtained based on the change of the oxygen adsorption characteristic due to the secular change of the lean NOx catalyst 4,
The correction control on the rich side of / F becomes excessive, and as a result,
There is no possibility of causing rich spikes and undesirably increasing CO and HC in the exhaust gas.

Further, in the lean burn engine system of the embodiment, the rear oxygen sensor 7 is arranged downstream of the catalytic converter 3. The rear oxygen sensor 7 and the lean NOx catalyst 4 in the catalytic converter 3 are combined with each other. You may make it arrange | position between the original catalysts 5. Further, in the catalytic converter 3, instead of the three-way catalyst, another catalyst for purifying NOx when the A / F is at the stoichiometric air-fuel ratio may be used.

[0042]

As described above, according to the air-fuel ratio control method of the lean burn engine system of the present invention, when the air-fuel ratio control is switched from the lean control mode to the stoichiometric air-fuel ratio control mode, the rear oxygen sensor Sensor signal is A
The air-fuel ratio is controlled on the rich side from the lean side to the stoichiometric air-fuel ratio side in terms of / F, so purification of NOx in the catalytic converter due to oxygen released from the lean NOx catalyst. It is possible to prevent a decrease in performance, and as a result, whether the air-fuel ratio is controlled in the lean control mode or the stoichiometric air-fuel ratio control mode,
Ox can be efficiently purified, and when the sensor signal of the rear oxygen sensor is reversed from the lean side, the control on the rich side of the air-fuel ratio is immediately stopped.
This has the effect of not causing deterioration of fuel efficiency.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a lean burn engine system.

FIG. 2 is a flowchart showing a part of an air-fuel ratio control routine.

FIG. 3 is a flowchart showing details of one step in FIG.

FIG. 4 is a flowchart showing a part of the air-fuel ratio control routine.

FIG. 5 is a flowchart showing a part of the air-fuel ratio control routine.

FIG. 6 is a graph showing a change over time of a rich gain.

FIG. 7 is a graph showing NOx purification rate characteristics in a catalytic converter when the air-fuel ratio control routine is not executed.

[Explanation of symbols]

 1 Engine 2 Exhaust Gas Pipe 3 Catalytic Converter 4 Lean NOx Catalyst 5 Three-Way Catalyst 6 Front Oxygen Sensor 7 Rear Oxygen Sensor 8 Electronic Control Unit (ECU) 10 Fuel Injector

Claims (1)

Claims: 1. A catalytic converter arranged in an exhaust gas pipe extending from an engine, and a lean NOx arranged in the exhaust gas pipe upstream of a catalyst of the catalytic converter.
A catalyst, a front oxygen sensor located in the exhaust gas pipe upstream of the lean NOx catalyst, for detecting the oxygen concentration in the exhaust gas to control the air-fuel ratio, and in the exhaust gas pipe than the lean NOx catalyst. A rear oxygen sensor, which is located downstream and is used to correct the air-fuel ratio control by detecting the oxygen concentration in the exhaust gas, and based on the sensor signals from the front and rear oxygen sensors, depending on the operating conditions of the engine. A lean burn engine system is provided with a control means for switching and controlling an air-fuel ratio of an engine between a lean control mode for controlling on the lean side and a stoichiometric air-fuel ratio control mode for controlling by a stoichiometric air-fuel ratio. When the fuel ratio is switched from the lean control mode to the stoichiometric air-fuel ratio control mode, the oxygen concentration in the exhaust gas detected by the rear oxygen sensor is lean when viewed from the air-fuel ratio. Until switched to Luo stoichiometric air-fuel ratio side, the air-fuel ratio control method for a lean-burn engine systems and controlling the air-fuel ratio of the engine in the rich side.