JP3899658B2 - Exhaust gas purification device for internal combustion engine - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an exhaust gas purification apparatus for an internal combustion engine, and more particularly to a technique for controlling an optimal air-fuel ratio necessary for NOx purification in an internal combustion engine having an NOx storage catalyst in an exhaust passage.
[0002]
[Prior art]
Conventionally, a NOx storage catalyst that stores NOx in exhaust when the exhaust air-fuel ratio is lean and releases the stored NOx when the exhaust air-fuel ratio is rich or stoichiometric or rich is reduced. An engine equipped with (NOx occlusion type three-way catalyst) is known.
[0003]
The NOx storage catalyst reduces the amount of NOx stored in the atmosphere by storing NOx during lean combustion, but if the stored amount exceeds the maximum amount, NOx discharged from the engine is stored in the catalyst. Without being discharged into the atmosphere.
Therefore, before the NOx occlusion amount in the NOx occlusion catalyst reaches the maximum amount, the target air-fuel ratio of the combustion mixture is forcibly temporarily switched to rich (hereinafter referred to as rich spike) and occluded in the NOx occlusion catalyst. NOx is released and reduced.
[0004]
Here, the NOx desorption characteristic of the NOx occlusion catalyst will be described. When rich exhaust gas is supplied by the rich spike, the occluded NOx is initially desorbed in large quantities and then gradually desorbed. Has characteristics. For this reason, control is performed such that the rich level at the beginning of the control is increased in accordance with the desorption characteristics, and then the rich level is decreased and the rich state is continued for a predetermined time.
[0005]
By the way, the NOx desorption characteristic of the NOx storage catalyst varies depending on the temperature of the catalyst. When the temperature is high, the initial desorption amount immediately after the start of the rich spike is large and the desorption is completed in a short time. It decreases and the desorption end time is prolonged.
Therefore, a technique for controlling the rich level and rich duration of the rich spike so as to adapt to different NOx desorption characteristics depending on the catalyst temperature has been proposed (see JP-A-6-10725).
[0006]
[Problems to be solved by the invention]
However, the NOx desorption characteristics change not only with the catalyst temperature, but also with the deterioration state and variation of the catalyst, the operating state of the engine, etc., and the control corresponding only to the catalyst temperature cannot cope with these changes.
The present invention has been made in view of the above problems, and even when the NOx desorption characteristic changes due to deterioration, variation or engine operating state of the NOx storage catalyst, it is always optimal for NOx purification according to the change in the characteristic. An object of the present invention is to provide an exhaust purification device that can be controlled to rich spike characteristics, and that can efficiently suppress NOx and suppress HC and CO emissions.
[0007]
[Means for Solving the Problems]
Therefore, the invention according to claim 1
A NOx storage catalyst that stores NOx in the exhaust when the exhaust air-fuel ratio is lean, and releases the stored NOx when the exhaust air-fuel ratio is rich or rich and performs a reduction process, and a combustion mixture In an exhaust gas purification apparatus for an internal combustion engine that performs control to reduce NOx stored in the NOx storage catalyst by temporarily setting the air-fuel ratio of the engine to be rich,
Wherein detecting a NOx storage catalyst downstream of the NOx concentration before and after the NOx reduction treatment control, based on the detected NOx concentration was, learning and rich level of the air-fuel ratio of the combustion mixture that definitive during NOx reduction treatment control, combustion mixture The rich continuation time learning is performed and corrected (see FIG. 1).
[0008]
According to the invention of claim 1,
When NOx reduction treatment control is performed to make the air-fuel ratio of the combustion mixture temporarily rich, the NOx emission amount from the NOx storage catalyst increases rapidly at the beginning of the control and decreases after the control from before the control. Depending on the air-fuel ratio control characteristics of the combustion mixture, the characteristics of the NOx emission amount at the initial stage of control and after the control change.
[0009]
Therefore, by detecting the NOx concentration before and after the NOx reduction treatment control and learning and correcting the air-fuel ratio control characteristic of the combustion mixture so as to match the characteristic of the NOx emission amount from the NOx storage catalyst based on the detected value, Sufficient NOx reduction treatment can be performed while suppressing the amount of HC emission .
[0012]
Also, for example, when the rich level of the air-fuel ratio of the combustion mixture is too low, NOx that is desorbed in a large amount at the beginning of the control due to insufficient HC emissions in the exhaust gas cannot be processed sufficiently, and conversely the rich level If it is too high, excessive amounts of HC and CO will be released. Therefore, the rich level can be appropriately learned and corrected based on the NOx concentration in the exhaust gas immediately after the start of the NOx reduction process control.
[0013]
Further, if the rich duration time is too short, NOx continues to be desorbed from the NOx storage catalyst even after the end of the control, and if the rich duration time is too long, the rich empty space will remain even after the NOx desorption from the NOx storage catalyst is finished. As combustion at the fuel ratio continues, HC and CO emissions increase. Accordingly, the rich duration time can be appropriately learned and corrected based on the NOx concentration in the exhaust gas after the end of the NOx reduction process control.
[0014]
The invention according to claim 2
The NOx reduction process control is a control in which the rich level of the air-fuel ratio of the combustion mixture is maximized at the start of control, and then gradually decreases. The rich level learning correction is performed at the maximum rich level given at the start of control. It is a learning correction of the level, and the learning correction of the rich duration time is a learning correction of the decrease rate of the rich level.
[0015]
According to such a configuration, the maximum rich level given at the start of the NOx reduction process control is learned and corrected by rich level learning, and then the speed at which the rich level is reduced is corrected by learning of the rich duration time.
As a result, the maximum rich level is learned and corrected to an appropriate value, so that the NOx emission amount at the start of the rich spike is suppressed while the HC and CO emission amounts are also suppressed. Is corrected to an appropriate value, so that the amount of NOx stored in the entire NOx storage catalyst can be sufficiently reduced while suppressing the emission of HC and CO.
[0016]
The invention according to claim 3
The learning of the air-fuel ratio control characteristic of the combustion mixture during the NOx reduction process control is performed for each engine operating region.
According to the invention of claim 3 ,
Even if the NOx concentration before and after the NOx reduction treatment control changes according to the engine operating conditions, it is possible to perform optimal learning according to the operating conditions by learning for each operating region.
[0017]
The invention according to claim 4
The air-fuel ratio control characteristic of the combustion mixture during the NOx reduction treatment control is also corrected by the temperature of the NOx storage catalyst.
According to the invention of claim 4 ,
Even if the NOx concentration before and after the NOx reduction treatment control changes depending on the temperature of the NOx storage catalyst, the optimum NOx reduction treatment control can be performed by performing correction according to the temperature of the catalyst.
[0018]
The invention according to claim 5
Learning of the air-fuel ratio control characteristic of the combustion mixture during the NOx reduction process control is prohibited when the engine operating condition is switched.
According to the invention of claim 5 ,
When the engine operating conditions are switched, the NOx concentration is changed accordingly. By prohibiting learning of the air-fuel ratio control characteristic of the combustion mixture based on the NOx concentration at the time of switching, erroneous learning can be prevented and learning is performed. Reliability is improved.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below.
FIG. 2 is a diagram showing a system configuration of the internal combustion engine in the first embodiment. The engine 1 is supplied with air measured by the throttle valve 2 and injected from the fuel injection valve 3 in the combustion chamber. The mixed gas is ignited and combusted by spark ignition by the spark plug 4, and the combustion exhaust gas is purified by the NOx storage catalyst 5 interposed in the exhaust passage 9 and then discharged into the atmosphere.
[0020]
The NOx occlusion catalyst 5 occludes NOx in the exhaust when the exhaust air-fuel ratio is lean, and releases the occluded NOx when the exhaust air-fuel ratio is rich in the stoichiometric air-fuel ratio or rich. It is a catalyst (NOx occlusion type three-way catalyst) for reduction treatment.
In addition, by directly injecting fuel into the combustion chamber as described above, it becomes possible to perform stratified mixture combustion with the air-fuel ratio made lean to the limit. In this case, NOx can be reduced while maintaining the lean air-fuel ratio. Since it is difficult, the NOx occlusion catalyst 5 temporarily occludes NOx, and the occluded NOx is reduced under a rich air-fuel ratio. However, lean combustion may be performed by injecting fuel into the intake port portion.
[0021]
A control unit 6 for controlling the injection timing and injection amount by the fuel injection valve 3 and the ignition timing by the spark plug 4 includes a microcomputer. The control unit 6 controls the fuel by arithmetic processing based on detection signals from various sensors. A fuel injection signal (injection pulse signal) is output to the injection valve 3, and an ignition signal is output to the spark plug 4 (power transistor).
[0022]
In the calculation of the fuel injection signal, the target air-fuel ratio is determined according to the operating conditions, and the fuel injection amount (injection pulse width) is calculated so that an air-fuel mixture of the target air-fuel ratio is formed. An air-fuel ratio that is leaner than the stoichiometric air-fuel ratio is set as the target air-fuel ratio.
The various sensors include an air flow meter 7 for detecting the intake air flow rate of the engine 1, a throttle sensor 8 for detecting the opening degree of the throttle valve 2, and an exhaust passage 9 upstream of the NOx storage catalyst 5. An air-fuel ratio sensor 10 that detects the air-fuel ratio, a NOx concentration sensor 11 that is disposed in the exhaust passage 9 on the downstream side of the NOx storage catalyst 5 and detects the NOx concentration, and a catalyst temperature that is disposed in the NOx storage catalyst 5 are detected. In addition to the catalyst temperature sensor 12 and the like, a rotation signal from the crank angle sensor 13 and a water temperature signal from the water temperature sensor 14 are input to the control unit 6.
[0023]
As described above, the control unit 6 normally sets the fuel injection amount so as to perform lean air-fuel ratio combustion, but using the NOx concentration sensor 11 while performing rich spike during NOx reduction processing control. While the NOx concentration downstream of the NOx occlusion catalyst 5 is detected, the control learning correction (hereinafter referred to as rich spike learning) is performed.
The NOx reduction process control according to the present invention will be described below with reference to FIGS.
[0024]
FIG. 3 shows a determination routine for performing the NOx reduction process control.
In step 1 (denoted as S1 in the figure, the same applies hereinafter), it is determined by the flag FLGRS whether or not the NOx reduction processing is being controlled.
When it is determined that the NOx reduction process control is not in progress, the routine proceeds to step 2 where the engine rotational speed Ne detected based on the signal from the crank angle sensor 13 and the engine rotational speed Ne and the air flow meter 7 are detected. The basic fuel injection amount Tp representing the engine load calculated based on the intake air amount Q is read.
[0025]
In step 3, based on the engine speed Ne and the basic fuel injection amount Tp, the NOx emission amount NOG per unit time (the execution cycle of this flowchart, for example, 1 second) is read from the characteristic map as shown in the figure. .
In step 4, the NOx emission amount NOG is added to the previous integrated value SIGNO to update the integrated value SIGNO.
[0026]
In step 5, it is determined whether or not the integrated value SIGNO of the NOx emission amount exceeds the set value SLSNO.
When it is determined that the set value SLSNO is exceeded, it is determined that the NOx stored in the NOx storage catalyst 5 has reached the state to be subjected to the reduction process, and the flag FRD, FLGRS and FLGFB are set to 1 respectively. The function of each flag will be described later.
[0027]
In step 7, the timer TRS for rich duration measurement is reset to zero. If it is determined in step 1 that NOx reduction processing is being controlled, the timer TRS is incremented in step 8, and the integrated value SIGNO is reset to 0 in step 9.
Next, a routine for setting the rich level and the rich duration in the NOx reduction process control will be described with reference to the flowchart of FIG.
[0028]
In step 11, based on the flag FRD, it is determined whether or not the NOx reduction process control execution condition has just been established.
If it is determined immediately after the establishment, the engine rotational speed Ne, the basic fuel injection amount Tp, and the catalyst temperature TCAT of the NOx storage catalyst 5 detected by the catalyst temperature sensor 12 are read in step 12.
[0029]
In step 13, the engine speed Ne, the basic fuel injection amount Tp, and the catalyst temperature TCAT are stored as TNe, TTp, and TTCAT, respectively, for rich spike learning described later.
In step 14, based on the engine speed Ne, the basic fuel injection amount Tp, and the catalyst temperature TCAT, a plurality of maps in which the initial value AFRRS of the rich level in the NOx reduction process control, that is, the rich air-fuel ratio is prepared for each temperature as shown in the figure. Search from and load. Here, the initial value AFRRS is such that the exhaust flow rate increases and the HC emission amount at the same air-fuel ratio increases as the engine speed Ne and the basic fuel injection amount Tp increase, so the air-fuel ratio is made thinner (from the theoretical air-fuel ratio). In addition, the higher the catalyst temperature TCAT, the greater the NOx desorption amount at the beginning of the rich spike, so the air-fuel ratio is set to be thicker. In the drawing, two maps for low temperature and high temperature of the catalyst temperature are shown, but three or more maps may be prepared for each temperature region.
[0030]
In step 15, based on the engine speed Ne, the basic fuel injection amount Tp, and the catalyst temperature TCAT, the rich continuation time TIMERS is retrieved and read from a plurality of maps prepared for each temperature as shown in the figure. Here, as the engine rotational speed Ne and the basic fuel injection amount Tp increase, the HC emission amount increases, so the duration TIMERS is lengthened. Also, as the catalyst temperature TCAT increases, the NOx desorption ends in a shorter time. The rich continuation time TIMERS is set short. Note that three or more maps may be prepared for each temperature region, as in the rich air-fuel ratio initial value map.
[0031]
In step 16, the flag FRD is reset to 0, the initial value AFRRS of the rich air-fuel ratio searched in steps 14 and 15 is stored as TAFRRS, and the rich duration TIMEERS is stored as TTIMERS. This is because these values are used for rich spike learning described later.
FIG. 5 shows an air-fuel ratio control routine with NOx reduction processing control, that is, rich spike characteristics.
[0032]
In step 21, the air-fuel ratio AFR being controlled is calculated by the following equation.
AFR = AFRRS + (14.6−AFRRS) / TIMERS × TRS
Here, TRS is the elapsed time after the start of control.
In step 22, it is determined whether or not the elapsed time TRS has reached the rich continuation time TIMERS. When the elapsed time TRS has been reached, that is, when the NOx reduction processing control has ended, the flag FLGRS is reset to 0 in step 23, A timer TIMER for learning described later is started.
[0033]
FIG. 6 shows changes in the air-fuel ratio during the NOx reduction process control. As shown in the figure, the air-fuel ratio AFR is controlled so as to be leaned at a constant speed so as to become the stoichiometric air-fuel ratio after the rich continuation time TIMERS has elapsed with the AFRRS as an initial value.
FIG. 7 shows a routine for determining whether or not rich spike learning is permitted according to the present invention.
[0034]
That is, at step 31, the change amount ΔTVO of the throttle valve opening TVO detected by the throttle sensor 8 is smaller than the reference value SLDTVO, or at step 32, the change amount ΔNe of the engine speed Ne is smaller than the reference value SLDNe. In 33, it is determined whether or not the engine coolant temperature TW detected by the water temperature sensor 14 is greater than the reference value SLTW. When all the conditions are YES, that is, the engine load and rotational fluctuation are small, and In the stable state that has been completed, learning is permitted. In other cases, there is no guarantee that the learning can be performed with high accuracy, so that the flag FLGFB is reset to 0 in step 34 to prohibit learning.
[0035]
FIG. 8 shows a routine for detecting the NOx concentration used for rich spike learning. Here, the NOx concentration sensor 11 and the routine of FIG. 8 constitute the NOx concentration detecting means according to the present invention.
In step 41, it is determined whether or not rich spike learning is permitted based on the value of the flag FLGFB.
[0036]
If the learning is permitted, the NOx concentration NO detected downstream of the NOx storage catalyst 5 detected by the NOx sensor 11 in step 42 is compared with a determination value NOMAX for maximum value detection, and the detected concentration NO is determined from the determination value NOMAX. When it is larger, the detected concentration NO is updated as a new NOMAX at step 43, and when the detected concentration NO is equal to or smaller than the determination value NOMAX, step 43 is jumped. By this operation, NOMAX finally left becomes the maximum value of the NOx concentration during the NOx detection period.
[0037]
In step 44, the detected concentration NO is compared with a determination value for detecting the minimum value. When the detected concentration NO is smaller than the determination value NOMIN, the detected concentration NO is updated as a new NOMIN in step 45, and the detected concentration NO is set. If it is greater than or equal to the determination value NOMAX, step 45 is jumped to. By this operation, NOMIN finally left becomes the minimum value of the NOx concentration during the NOx detection period.
[0038]
In step 46, it is determined whether the NOx concentration detection time measured by the timer TIMER is greater than the detection end determination value SLTIME. If it is greater than SLTIME, it is determined that the NOx concentration detection time has ended, and step 47 Thus, the ratios A and B of the maximum value NOMAX and the minimum value NOMIN to the NOP immediately before the start of the current NOx concentration detection (stored in step 49 as will be described later) are calculated by the following equations.
[0039]
A = NOMAX / NOP
B = NOMIN / NOP
In step 48, the value of the flag FLGFB is reset to zero.
When the value of the flag FLGFB is 0 in step 41, that is, when learning is prohibited, the values of NOMAX and NOMIN are reset to 0 in step 49, and the NOx concentration NO detected by the NOx sensor 12 is stored as NOP. Update. Thereby, as described above, the NOx concentration detected immediately before the next learning is used as the NOP.
[0040]
FIG. 9 shows the change in the NOx concentration after the start of the NOx reduction process control.
FIG. 10 shows a rich spike learning routine. The routine of FIG. 10 constitutes NOx reduction process control characteristic learning means according to the present invention.
In step 51, it is determined whether or not rich spike learning is permitted based on the value of the flag FLGFB.
[0041]
In step 52, it is determined whether the ratio A of the NOx concentration maximum value NOMAX calculated in step 47 of FIG. 8 to the immediately preceding value NOP is within the set range (SLAH>A> SLAL). If it is determined that the ratio A is within the set range, it is determined that it is not necessary to learn the rich level, and the routine jumps to step 57.
When it is determined that the ratio A is not within the set range, in step 53, based on the deviation of the ratio A from the center value [(SLAH + SLAL) / 2] of the set range, a rich level is determined from the map as shown in the figure. The correction value DAFR for learning is searched. Here, when the deviation is positive, the correction value DAFR is when the NOx desorption amount at the initial stage of the rich spike is too large. Therefore, the larger the deviation is, the richer the air-fuel ratio is set to increase the HC emission amount. If the correction value DAFR is a negative value and the absolute value is increased, and the deviation is negative, the amount of HC emission is excessive, so that the leaner the air-fuel ratio becomes, the larger the absolute value of the negative deviation becomes, and the HC The positive correction value DAFR is set to be increased so as to reduce the discharge amount.
[0042]
In step 54, the value (DAFR + TAFRRS) corrected by adding the correction value DAFR retrieved in step 54 to the initial value TAFRRS of the rich air-fuel ratio stored in step 16 of FIG. 4 is regulated by the upper and lower limit values. It is determined whether or not the value is within the range (SLLAFR <DAFR + TAFRRS <SLHAFR).
[0043]
When it is determined that the value is within the set range, the corrected value (DAFR + TAFRRS) is updated at step 55 as the initial value TAFRRS of the rich air-fuel ratio.
In step 56, based on the current engine speed TTNe, basic fuel injection amount TTp, and catalyst temperature TTCAT stored in step 13 of FIG. 4, the updated initial value TAFRRS is shown in the map shown in step 14 of FIG. The update is stored in the corresponding area.
[0044]
After step 57, the rich continuation time is learned in a similar manner under a predetermined condition. That is, in step 57, it is determined whether or not the ratio B of the NOx concentration minimum value NOMIN to the immediately preceding value NOP is within the set range (SLBH>B> SLBL). When it is determined that the ratio B is within the set range, it is determined that it is not necessary to learn the rich continuation time, and this routine is terminated.
[0045]
If it is determined that the ratio B is not within the set range, the rich continuation is continued from the map as shown in FIG. 58 based on the deviation of the ratio B from the center value [(SLBH + SLBL) / 2] of the set range. The correction value DTIME for time learning is searched. Here, since the rich continuation time is too short when the deviation is positive, the correction value DTIME is increased when the NOx is desorbed even after the rich spike ends. If the correction value DTIME is increased and the deviation is negative, the rich duration time is too long. Therefore, the larger the absolute value of the negative deviation is, the smaller the air duration ratio should be decreased. The absolute value of the negative correction value DTIME is set to be increased.
[0046]
At step 59, the rich continuation time TTIMER stored at step 16 of FIG. 4 is added to the correction value DTIME searched at step 58 to correct the value (DTIME + TIMER) that is regulated by the upper and lower limits (SLLTIME). <DTIME + TTIMES <SLHTIME) It is determined whether or not it is within the set range, and when it is out of the setting range, this routine is terminated without learning the rich duration time TTIMER and maintained at the current value.
[0047]
If it is determined that the value is within the set range, the corrected value (DTIME + TTIMER) in step 60 is updated as the rich duration time TTIMER.
In step 61, based on the current engine speed TTNe, basic fuel injection amount TTp, and catalyst temperature TTCAT, the updated rich duration time TTIMER is updated and stored in the corresponding area of the map shown in step 15 of FIG. To do.
[0048]
In this way, by learning and correcting the rich level and rich duration in the NOx reduction processing control based on the detection result of the NOx concentration, the NOx desorption characteristic of the NOx storage catalyst 5 depends on the deterioration, variation, and operating state of the catalyst. Even if there is a change, the air-fuel ratio control of the combustion mixture that is suitable for the NOx desorption characteristic can be performed, and the NOx amount stored in the entire NOx storage catalyst can be sufficiently reduced while suppressing the emission of HC and CO. it can.
[0049]
In FIG. 5, instead of the initial value AFRRS and rich duration time TIMER of the rich air-fuel ratio, the NOx reduction process control may be executed using the initial value TAFRRS and rich duration time TTIMER that have been learned in FIG. 10. Well, the learning result can be reflected in the control one time earlier.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a basic configuration of an exhaust emission control device according to a second aspect of the invention.
FIG. 2 is a system configuration diagram of the internal combustion engine in the embodiment.
FIG. 3 is a flowchart showing a determination routine for performing NOx reduction process control in the embodiment;
FIG. 4 is a flowchart showing a routine for setting a rich level and a rich continuation time in NOx reduction processing control according to the embodiment;
FIG. 5 is a flowchart showing a NOx reduction process control routine in the embodiment.
FIG. 6 is a diagram showing a change in air-fuel ratio during rich spike air-fuel ratio control in the embodiment.
FIG. 7 is a flowchart showing a routine for determining whether or not rich spike learning is permitted in the embodiment;
FIG. 8 is a flowchart showing a routine for detecting a NOx concentration used for rich spike learning in the embodiment;
FIG. 9 is a time chart showing a change in NOx concentration after the start of NOx reduction treatment control in the embodiment.
FIG. 10 is a flowchart showing a rich spike learning routine in the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Internal combustion engine 2 Throttle valve 3 Fuel injection valve 5 NOx storage catalyst 6 Control unit 7 Air flow meter 8 Throttle sensor 9 Exhaust passage
11 NOx concentration sensor
12 Catalyst temperature sensor
13 Crank angle sensor
14 Water temperature sensor

Claims (5)

A NOx storage catalyst that stores NOx in the exhaust when the exhaust air-fuel ratio is lean, and releases the stored NOx when the exhaust air-fuel ratio is rich or rich and performs a reduction process, and a combustion mixture In an exhaust gas purification apparatus for an internal combustion engine that performs control to reduce NOx stored in the NOx storage catalyst by temporarily setting the air-fuel ratio of the engine to be rich,
Wherein detecting a NOx storage catalyst downstream of the NOx concentration before and after the NOx reduction treatment control, based on the detected NOx concentration was, learning and rich level of the air-fuel ratio of the combustion mixture that definitive during NOx reduction treatment control, combustion mixture The exhaust gas purifying device for an internal combustion engine, wherein the rich continuation time is learned and corrected . The NOx reduction process control is a control in which the rich level of the air-fuel ratio of the combustion mixture is maximized at the start of control, and then gradually decreases. The rich level learning correction is performed at the maximum rich level given at the start of control. 2. The exhaust gas purification apparatus for an internal combustion engine according to claim 1 , wherein the learning correction of the rich duration is a learning correction of the decrease rate of the rich level. 3. The exhaust purification device for an internal combustion engine according to claim 1 or 2, wherein learning of the air-fuel ratio control characteristic of the combustion air-fuel mixture during the NOx reduction process control is performed for each engine operating region. The exhaust of the internal combustion engine according to any one of claims 1 to 3 , wherein the air-fuel ratio control characteristic of the combustion mixture during the NOx reduction treatment control is also corrected by the temperature of the NOx storage catalyst. Purification equipment. The internal combustion engine according to any one of claims 1 to 4 , wherein learning of the air-fuel ratio control characteristic of the combustion mixture during the NOx reduction treatment control is prohibited when the operating condition of the engine is switched. Engine exhaust purification system.

Images