JP2005351250A - Air fuel ratio control device for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To always maintain the best emission characteristics in a downstream of a catalyst in relation to air fuel ratio control device for an internal combustion engine. <P>SOLUTION: A main air fuel ratio sensor and a sub O<SB>2</SB>sensor are arranged in an upstream of a catalyst arranged in an exhaust gas passage and in a downstream thereof respectively. Fuel injection quantity is controlled based on output of the main air fuel ratio sensor and output of the sub O<SB>2</SB>sensor. Fuel injection quantity is controlled to make air fuel ratio in the upstream of the catalyst richer than the actual stoichiometric air fuel ratio of the catalyst (straight line 2) in a zone where intake air quantity Ga exceeds 15 g/sec. The enrichment tendency is established to obtain the overall best emission characteristics. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an air-fuel ratio control apparatus for an internal combustion engine, and more particularly to an air-fuel ratio control apparatus suitable as an apparatus for controlling the air-fuel ratio of an on-vehicle internal combustion engine.

  2. Description of the Related Art Conventionally, as disclosed in, for example, Japanese Patent Application Laid-Open No. 7-197837, an internal combustion engine is known that includes two exhaust gas sensors in an exhaust passage of the internal combustion engine. This internal combustion engine includes an air-fuel ratio sensor (a sensor that exhibits linear characteristics with respect to the air-fuel ratio) upstream of the catalyst disposed in the exhaust passage, and an oxygen sensor (so-called Z characteristics with respect to the air-fuel ratio) downstream of the catalyst. Sensor).

  In the conventional internal combustion engine, the main feedback control is executed based on the output of the upstream air-fuel ratio sensor, while the sub feedback control is executed based on the output of the downstream oxygen sensor. In the main feedback control, the fuel injection amount is controlled so that the air-fuel ratio of the exhaust gas flowing into the catalyst matches the control target air-fuel ratio. In the sub-feedback control, more specifically, the output of the oxygen sensor arranged downstream of the catalyst is the stoichiometric output so that the air-fuel ratio of the exhaust gas flowing out downstream of the catalyst becomes the stoichiometric air-fuel ratio. Thus, the content of the main feedback control is corrected. According to these controls, the air-fuel ratio downstream of the catalyst can be systematically maintained at a value close to the theoretical air-fuel ratio, and excellent emission characteristics can be realized.

JP-A-7-197837 Japanese Patent Laid-Open No. 9-60544

  In the above-described conventional internal combustion engine, in principle, control for changing the output of the oxygen sensor downstream of the catalyst to the stoichiometric output is continuously performed. However, the present inventor has found that the best emission characteristics are not necessarily obtained only when the oxygen sensor downstream of the catalyst emits a stoichiometric output. More specifically, the inventor of the present application emits an output biased to the rich side rather than the case where the oxygen sensor downstream of the catalyst emits the stoichiometric output under the situation where the intake air amount Ga is large. It was found that excellent emission characteristics can be obtained stably when

  According to the above knowledge, it is desirable to control the fuel injection amount so that the oxygen sensor downstream of the catalyst emits a richly biased output under a situation where the intake air amount Ga is large. In this regard, the control executed in the conventional internal combustion engine cannot always achieve the best emission characteristics.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an air-fuel ratio control apparatus for an internal combustion engine that can always maintain the best emission characteristics downstream of the catalyst.

In order to achieve the above object, a first invention is an air-fuel ratio control apparatus for an internal combustion engine,
A catalyst disposed in an exhaust passage of the internal combustion engine;
An upstream exhaust gas sensor disposed upstream of the catalyst;
A downstream side exhaust gas sensor disposed downstream of the catalyst;
Intake air amount detection means for detecting the intake air amount of the internal combustion engine;
Fuel injection amount control means for controlling the fuel injection amount based on the output of the upstream side exhaust gas sensor and the output of the downstream side exhaust gas sensor;
The fuel injection amount control means adjusts the fuel injection amount so that the air-fuel ratio upstream of the catalyst becomes a predetermined air-fuel ratio richer than the actual stoichiometric air-fuel ratio of the catalyst in a region where the intake air amount exceeds a predetermined value. Control,
The predetermined air-fuel ratio is a value determined in relation to the amount of intake air, in order to realize a desired emission characteristic.

In a second aspect based on the first aspect, the fuel injection amount control means comprises:
Means for calculating a feedback correction basic amount for setting the air-fuel ratio upstream of the catalyst to the target air-fuel ratio based on the output of the upstream side exhaust gas sensor;
Means for calculating a sub feedback correction amount for matching them based on the difference between the output of the downstream side exhaust gas sensor and the stoichiometric output;
Catalyst characteristic correction amount calculating means for setting a catalyst characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the predetermined air-fuel ratio based on the intake air amount;
A corrected correction amount calculating means for calculating a corrected feedback correction amount by correcting the feedback correction basic amount based on the sub feedback correction amount and the catalyst characteristic correction amount;
Means for calculating the fuel injection amount based on the post-correction feedback correction amount;
It is characterized by providing.

In a third aspect based on the second aspect, the fuel injection amount control means comprises:
In the region where the intake air amount is less than or equal to the predetermined value, the fuel injection amount is controlled so that the air-fuel ratio upstream of the catalyst becomes the actual stoichiometric air-fuel ratio of the catalyst, and
Means for calculating a sub feedback learning amount by taking a steady value included in the sub feedback correction amount;
A corrected correction amount calculating means for calculating the feedback correction basic amount based on the sub feedback learning amount in addition to the sub feedback correction amount and the catalyst characteristic correction amount;
It further comprises learning region limiting means for permitting the update of the sub feedback learning amount only when the intake air amount is equal to or less than the predetermined value.

In a fourth aspect based on the first aspect, the fuel injection amount control means comprises:
Means for calculating a feedback correction basic amount for setting the air-fuel ratio upstream of the catalyst to the target air-fuel ratio based on the output of the upstream side exhaust gas sensor;
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Sub feedback correction amount calculation means for calculating a value obtained by eliminating the influence of the rich correction amount as a sub feedback correction amount from a correction amount for matching the output of the downstream exhaust gas sensor and the stoichiometric output;
A corrected correction amount calculating means for calculating a corrected feedback correction amount by correcting the feedback correction basic amount based on the characteristic correction amount, the rich correction amount, and the sub feedback correction amount;
Means for calculating the fuel injection amount based on the post-correction feedback correction amount;
It is characterized by providing.

Further, in a fifth invention according to any one of the second to fourth inventions, the catalyst characteristic correction amount calculating means comprises:
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Means for calculating the catalyst characteristic correction amount by adding the characteristic correction amount and the rich correction amount;
Means for detecting an output upper limit value of the downstream side exhaust gas sensor;
Means for calculating a rich correction amount upper limit guard value smaller than the output upper limit value based on the output upper limit value;
Means for replacing the rich correction amount with the rich correction amount upper limit guard value when the rich correction amount is larger than the rich correction amount upper limit guard value;
Is further provided.

Further, in a sixth invention according to any one of the second to fifth inventions, the catalyst characteristic correction amount calculating means comprises:
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Means for calculating the catalyst characteristic correction amount by adding the characteristic correction amount and the rich correction amount;
Means for detecting an output lower limit value of the downstream side exhaust gas sensor;
Means for calculating a rich correction amount lower limit guard value larger than the output upper limit value based on the output lower limit value;
Means for replacing the rich correction amount with the rich correction amount lower limit guard value when the rich correction amount is larger than the rich correction amount guard value;
Is further provided.

  According to the first aspect of the invention, in the region where the intake air amount Ga exceeds the predetermined value, the air-fuel ratio upstream of the catalyst is determined based on the output of the upstream exhaust gas sensor and the output of the downstream exhaust gas sensor. The fuel injection amount can be controlled so as to be a predetermined air-fuel ratio richer than the fuel ratio (the air-fuel ratio upstream of the catalyst that can make the air-fuel ratio downstream of the catalyst the theoretical air-fuel ratio). The predetermined air-fuel ratio is a value determined in advance to realize desired emission characteristics under a situation where the intake air amount Ga is large. If the air-fuel ratio upstream of the catalyst is controlled to such a predetermined air-fuel ratio, good emission characteristics can be obtained more stably than when the air-fuel ratio is the actual stoichiometric air-fuel ratio. For this reason, according to the present invention, it is possible to stably achieve good emission characteristics in a situation where the intake air amount Ga is large.

  According to the second aspect, the corrected feedback correction amount can be calculated by correcting the feedback correction basic amount based on the sub-feedback correction amount and the catalyst characteristic correction amount. The catalyst characteristic correction amount is a correction amount for setting the air-fuel ratio upstream of the catalyst to a predetermined air-fuel ratio. Therefore, when the feedback correction basic amount is corrected with the catalyst characteristic correction amount, the fuel injection amount can be controlled so that the air-fuel ratio upstream of the catalyst becomes the predetermined air-fuel ratio. The sub feedback correction amount is a correction amount for making the output of the downstream side exhaust gas sensor a stoichiometric output. For this reason, if the feedback correction basic amount is corrected with the sub feedback correction amount, it is possible to prevent the air-fuel ratio downstream of the catalyst from being excessively rich with respect to the predetermined air-fuel ratio. For this reason, according to the present invention, it is possible to stably realize desired emission characteristics in a situation where the intake air amount Ga is large.

  According to the third invention, in the region where the intake air amount Ga is small, the air-fuel ratio upstream of the catalyst can be made the actual stoichiometric air-fuel ratio of the catalyst. In this case, the air-fuel ratio downstream of the catalyst should be stoichiometric. Therefore, the sub-feedback correction amount is updated to a value mainly corresponding to the deviation inherent in the system. According to the present invention, the update of the sub feedback learning amount is allowed only under such a situation. For this reason, according to the present invention, it is possible to reliably prevent the sub-feedback learning amount from being updated to an inappropriate value, that is, a value that does not have a close correlation with the deviation inherent in the system.

  According to the fourth aspect of the invention, the corrected feedback correction amount can be calculated by correcting the feedback correction basic amount based on the characteristic correction amount, the rich correction amount, and the sub feedback correction amount. In the present invention, the sub feedback correction amount is obtained by eliminating the influence of the rich correction amount from the correction amount for setting the output of the downstream side exhaust gas sensor to the stoichiometric output. In this case, only the influence of the deviation inherent in the system is reflected in the sub feedback correction amount. Therefore, according to the present invention, the sub-feedback learning amount is constantly updated to a value having a close correlation with the deviation inherent in the system, while the upstream air-fuel ratio is enriched by the correction using the characteristic correction amount and the rich correction amount. can do.

  According to the fifth aspect of the invention, the characteristic correction amount for making the air-fuel ratio upstream of the catalyst the actual stoichiometric air-fuel ratio of the catalyst and the air-fuel ratio upstream of the catalyst are rich by the difference between the actual stoichiometric air-fuel ratio and the predetermined air-fuel ratio. The sum of the correction amount and the rich correction amount can be used as the catalyst characteristic correction amount. Here, the rich correction amount can be guarded to a rich correction amount upper limit guard value smaller than the output upper limit value of the downstream side exhaust sensor. Therefore, according to the present invention, it is possible to reliably prevent a situation in which the output of the downstream side exhaust gas sensor sticks to the output upper limit value due to the correction by the catalyst characteristic correction amount.

  According to the sixth aspect of the invention, the rich correction amount can be guarded to the rich correction amount lower limit guard value that is larger than the output lower limit value of the downstream side exhaust sensor. Therefore, according to the present invention, it is possible to reliably prevent a situation in which the output of the downstream side exhaust gas sensor sticks to the output lower limit value due to the correction by the catalyst characteristic correction amount.

Embodiment 1 FIG.
[Configuration of Device of Embodiment 1]
FIG. 1 is a diagram for explaining the configuration of an air-fuel ratio control apparatus according to Embodiment 1 of the present invention. As shown in FIG. 1, the apparatus of this embodiment includes an upstream catalyst (S / C) 12 and a downstream catalyst (U / F) 14 disposed in an exhaust passage 10 of the internal combustion engine. The upstream catalyst 12 and the downstream catalyst 14 are all three-way catalysts that can simultaneously purify CO, HC, and Nox.

A main air-fuel ratio sensor 16 and a sub O ₂ sensor 18 are disposed upstream and downstream of the upstream catalyst 12, respectively. The main air-fuel ratio sensor 16 is a sensor that generates an output evafbse that is substantially linear with respect to the air-fuel ratio A / F of the exhaust gas flowing into the upstream catalyst 12. On the other hand, the sub O ₂ sensor 18 generates a rich output (for example, 0.8 V) when the exhaust gas flowing out from the upstream catalyst 12 is rich with respect to the stoichiometric air-fuel ratio, and the exhaust gas is lean. In this case, the sensor generates a lean output (for example, 0.2 V).

The output evafbse of the main air-fuel ratio sensor 16 and the output voxs of the sub O ₂ sensor 18 are respectively supplied to an ECU (Electronic Control Unit) 20. The ECU 20 is further connected to an air flow meter 22, a rotation speed sensor 24, a fuel injection valve 26, and the like. The air flow meter 22 is a sensor that detects an intake air amount Ga of the internal combustion engine. The rotational speed sensor 24 is a sensor that generates an output corresponding to the engine rotational speed Ne. The fuel injection valve 26 is an electromagnetic valve for injecting fuel into the intake port of the internal combustion engine.

[Specific Processing in Embodiment 1]
Next, the contents of specific processing executed in the system of this embodiment will be described with reference to FIG. FIG. 2 shows a flowchart of a routine executed by the ECU 20 in the present embodiment.

  In the routine shown in FIG. 2, first, the output evafbse (mV) of the main air-fuel ratio sensor 16 is captured (step 100).

  Next, processing relating to the stoichiometric point learning amount evafofs (mV) is executed (step 102). The stoichiometric point learning is learning for correcting the influence of variations in the wire harness on the output evafbse of the main air-fuel ratio sensor 16. In step 120, processing for updating the stoichiometric point learning amount evafofs is executed on condition that a predetermined stoichiometric point learning condition is satisfied. Note that the calculation method of the stoichiometric point learning amount evafofs is not a main part of the present invention, and therefore detailed description thereof is omitted here.

Next, the sub feedback correction amount evafsfb (mV) is calculated (step 104). Specifically, on the condition that the sub feedback execution condition is satisfied, the deviation dlvoxs = voxsref−voxs between the output voxs of the sub O ₂ sensor 18 and the stoichiometric output voxsref (0.5 V) is Further, based on the deviation dlvoxs, the sub feedback correction amount evafofs is calculated by PID control, for example.

  Next, it is determined whether or not the intake air amount Ga is equal to or less than a predetermined determination value (15 g) (step 106). That is, it is determined whether or not the intake air amount Ga is small.

  If it is determined that Ga ≦ 15g is established as a result of the above determination, next, processing relating to the sub feedback learning amount evafsfbg (mV) is executed (step 108). Here, specifically, at a predetermined timing, the process of transferring the average value of the sub feedback correction amount evafsfb to the sub feedback learning amount evafsfbg, that is, the sub feedback learning amount evafsfb is increased or decreased by the average value, In addition, processing for decreasing or increasing the sub feedback correction amount evafsfb by the average value is performed.

  On the other hand, if the establishment of Ga ≦ 15g is denied in step 106, it is determined that learning of the sub-feedback learning amount evafsfbg should not proceed, and the processing of step 108 is jumped.

  Next, processing relating to the catalyst characteristic correction amount evafcat (mV) is executed (step 110). FIG. 3 is a map that defines the relationship between the intake air amount Ga and the catalyst characteristic correction amount evafcat. In step 108, the ECU 20 refers to this map to calculate the catalyst characteristic correction amount evafcat corresponding to the current intake air amount Ga. The details of FIG. 3 and the catalyst characteristic correction amount evafcat will be described in detail later.

In the routine shown in FIG. 2, the corrected A / F output evabyf (mV) is then calculated according to the following arithmetic expression (step 112).
evabyf = evafbse + evafofs + evafsfb + evafsfbg + evafcat (1)

  Next, the control target air-fuel ratio eabyfref is calculated based on the operating state of the internal combustion engine 10 (step 114). For example, when the operation at the stoichiometric air-fuel ratio is required, the control target air-fuel ratio is set to 14.6.

  Finally, fuel injection amount control for making the corrected A / F output evabyf (mV) correspond to the control target air-fuel ratio eabyfref, that is, main feedback control is executed (step 116). Specifically, first, the corrected A / F output evabyf (mV) is first converted into an air-fuel ratio. When the air-fuel ratio obtained by the conversion is larger than the control target air-fuel ratio eabyfref (lean), the fuel injection amount is increased and corrected by a value corresponding to the difference between the two. When the air-fuel ratio obtained by the conversion is smaller than the control target air-fuel ratio eabyfref (rich), the fuel injection amount is corrected to be reduced by a value corresponding to the difference between the two.

[Features of Embodiment 1]
Next, characteristics of the first embodiment of the present invention will be described with reference to FIGS. As described above, the system of the present embodiment executes main feedback control for making the corrected A / F output evabyf (mV) correspond to the control target air-fuel ratio eabyfref (see step 116 above). Then, the corrected A / F output evabyf (mV) is calculated according to the above equation (1).

  According to the above equation (1), when calculating the corrected A / F output evabyf, the stoichiometric learning amount evafofs is added to the output evafbse of the main air-fuel ratio sensor 16. According to the stoichiometric learning amount evafofs, it is possible to cancel out the influence of the wire harness and the like, and to accurately cancel out the error superimposed on the output of the main air-fuel ratio sensor 16. Therefore, the value “evafbse + evafofs” is a value that accurately represents the air-fuel ratio upstream of the upstream catalyst 12 (hereinafter referred to as “upstream air-fuel ratio”).

According to the above equation (1), in calculating the corrected A / F output evabyf, the sub feedback correction amount evafsfb is added to the output evafbse of the main air-fuel ratio sensor 16. As described above, the sub feedback correction amount evafsfb is a value calculated based on the deviation dlvoxs = voxsref−voxs between the output voxs of the sub O ₂ sensor 18 and the stoichiometric output voxsref. The deviation dlvoxs is negative when the output voxs of the sub O ₂ sensor 18 is a rich output (0.8 V), while it is positive when the output voxs is a lean output (0.2 V). For this reason, the sub-feedback correction amount evafsfb is normally a negative value when the sub O ₂ sensor 18 emits a rich output, and is a positive value when the sub O ₂ sensor 18 emits a lean output.

The output voxs of the sub O ₂ sensor 18 is rich when the upstream air-fuel ratio is biased to be rich, and as a result, rich exhaust gas is blown downstream of the upstream catalyst 12. According to the negative sub-feedback correction amount evafsfb, the corrected A / F output evabyf can be corrected to a small value. When the corrected A / F output evabyf is corrected to a small value, that is, a rich value, the content of the main feedback control is corrected to reduce the fuel injection amount. Therefore, according to the sub feedback correction amount evafsfb described above, when the upstream air-fuel ratio is biased to the rich side, the bias can be corrected.

Based on the same principle, according to the sub feedback correction amount evafsfb described above, even when the upstream air-fuel ratio is biased toward the lean side, the bias can be corrected. Thus, by reflecting the sub-feedback correction amount evafsfb in the corrected A / F output evabyf, the air-fuel ratio downstream of the upstream catalyst 12 (hereinafter referred to as “downstream air-fuel ratio”) is maintained at the stoichiometric air-fuel ratio. More precisely, the content of the main feedback control can be appropriately corrected so that the state in which the sub O ₂ sensor 18 can generate the stoichiometric output is maintained. Hereinafter, correcting the content of the main feedback control as described above is referred to as “sub-feedback control”.

  According to the above equation (1), in calculating the corrected A / F output evabyf, the catalyst characteristic correction amount evafcat is added to the output evafbse of the main air-fuel ratio sensor 16. As described above, the catalyst characteristic correction amount evafcat is a value set in relation to the intake air amount Ga with reference to the map shown in FIG. Before explaining the contents of the map shown in FIG. 3, the reason why the catalyst characteristic correction amount evafcat is set as shown in FIG. 3 will be described with reference to FIGS.

(Description of catalyst characteristic correction amount evafcat)
FIG. 4 is a diagram for explaining the four straight lines (1) to (4) that are examined in creating the map of the catalyst characteristic correction amount evafcat. In FIG. 4, the vertical axis indicates the catalyst characteristic correction amount evafcat (mV) and its A / F conversion value. Further, the horizontal axis of FIG. 4 represents the intake air amount Ga (g / sec).

4 is a catalyst window of the upstream catalyst 12 (however, in this case, the vertical axis of FIG. 4 is regarded as an air-fuel ratio indicated as an A / F conversion value). To do). The catalyst window is the range of the air-fuel ratio at which the gas flowing out downstream of the upstream catalyst 12 becomes stoichiometric, more precisely, the range of the upstream air-fuel ratio in which the sub O ₂ sensor 18 can generate the stoichiometric output.

As shown in FIG. 4, the catalyst window shifts to the rich side as the intake air amount Ga increases. For this reason, in order to always maintain the air-fuel ratio of the exhaust gas flowing out from the upstream catalyst 12 at stoichiometric, that is, to always maintain the output voxs of the sub O ₂ sensor 18 at stoichiometric output, the intake air amount Ga increases. Accordingly, it is necessary to enrich the upstream air-fuel ratio.

  A straight line denoted by reference numeral (2) in FIG. 4 is a straight line set so as to pass through substantially the center of the catalyst window. Hereinafter, the air-fuel ratio that satisfies the relationship of the straight line (2) is referred to as the “actual theoretical air-fuel ratio” of the upstream catalyst 12.

  More precisely, this straight line (2) is a straight line that defines a catalyst characteristic correction amount evafcat for maintaining the upstream air-fuel ratio at the actual stoichiometric air-fuel ratio of the upstream catalyst 12 in relation to the intake air amount Ga. According to the straight line (2), the catalyst characteristic correction amount evafcat becomes zero when the intake air amount Ga is sufficiently small (when it is substantially zero), and increases proportionally as the intake air amount Ga increases. If the catalyst characteristic correction amount evafcat is a y value and the intake air amount Ga is an x value, the relationship between them can be expressed as y = 0.8x.

  If the catalyst characteristic correction amount evafcat is determined according to the relationship of the straight line (2), when the intake air amount Ga is almost zero, the corrected A / F output evabyf is a value obtained by adding zero to “evafbse + evafofs + evafsfb + evafsfbg” (Case 1). When the intake air amount Ga is about 40 (g / sec), the corrected A / F output evabyf is a value obtained by adding 32 (mV) to “evafbse + evafofs + evafsfb + evafsfbg” (case 2) (both 1) Refer to equation).

  That is, when the state transition from case 1 to case 2 occurs as the intake air amount Ga increases rapidly, the corrected A / F output evabyf is instantaneously set to 32 (mV without waiting for the follow-up of the sub feedback correction amount evafsfb. ) Will change to a larger value. If the control target air-fuel ratio is the same value, when the corrected A / F output evabyf increases, the fuel injection amount is increased by the function of the main feedback control. The increase is performed so that 32 (mV) generated by the addition of the catalyst characteristic correction amount evafcat is absorbed by the decrease in the output evafbse of the main air-fuel ratio sensor 16. As a result, according to the above example, with the transition from case 1 to case 2, the upstream air-fuel ratio is enriched by 32 (mV) added as the catalyst characteristic correction amount evafcat.

  The relationship between the catalyst characteristic correction amount evafcat and the A / F converted value indicated on the vertical axis in FIG. 4 is realized as a result of the catalyst characteristic correction amount evafcat used for calculating the corrected A / F output evabyf and its use. This corresponds to the relationship with the upstream air-fuel ratio. According to this relationship, when 32 (mV) is used as the catalyst characteristic correction amount evafcat, the upstream air-fuel ratio is predicted to be enriched to about 14.50.

  As described above, if the corrected A / F output evabyf is corrected using the catalyst characteristic correction amount evafcat, the control target air-fuel ratio is maintained at a constant value as the intake air amount Ga changes, The upstream air-fuel ratio can be changed appropriately. If the catalyst characteristic correction amount evafcat is set according to the straight line (2), the upstream air-fuel ratio is always changed along the center of the catalyst window, that is, the actual stoichiometric air-fuel ratio of the upstream catalyst 12 is maintained. Is possible.

  By the way, the straight line denoted by reference numeral (1) in FIG. 4 is a straight line in which the rate of change of the catalyst characteristic correction amount evafcat with respect to the change of the intake air amount Ga is set to twice that of the straight line (2). . A straight line denoted by reference numeral (3) is a straight line whose rate of change is ½ that of the straight line (2). Furthermore, a straight line denoted by reference numeral (4) is a straight line in which the above change rate is zero. These straight lines (1), (3), and (4) are respectively “y = 1.6x” and “y = 0.4x” where the catalyst characteristic correction amount is y value and the intake air amount Ga is x value. And “y = 0”. Hereinafter, the slopes of these straight lines are referred to as “catalyst characteristic correction slopes”.

  FIG. 5 shows the results of an experiment confirming the relationship between the inclination of the catalyst characteristic correction and the HC and NOx emissions. More specifically, the results of measuring the amounts of HC and NOx discharged from the upstream catalyst 12 after setting the inclination of the catalyst characteristic correction to each of the straight lines (1) to (4) are summarized. It is.

  The results shown in FIG. 5 indicate that the HC emission amount (see ◯) slightly increases and the NOx emission amount (see △) rapidly decreases as the inclination of the catalyst characteristic correction increases. In other words, this result shows that when comparing the slope of the catalyst characteristic correction to the slope of the straight line (2) and the case of setting the slope of the straight line (1), the HC emission amount is much larger between the two. On the other hand, no difference was observed, and regarding the amount of NOx emissions, the former showed a marked advantage over the latter.

  During the operation of the internal combustion engine, it is inevitable that a certain amount of air-fuel ratio roughening occurs. In order to obtain good emission characteristics under such a premise, that is, in order to reduce the HC emission amount and the NOx emission amount as a whole, it is effective to place importance on stability against air-fuel ratio roughening. If the stability is emphasized, it can be concluded that the inclination of the catalyst characteristic correction is preferably set to the slope of the straight line (1) rather than the slope of the straight line (2).

  Therefore, in the system of the present embodiment, in principle, the air-fuel ratio control is performed by setting the catalyst characteristic correction amount evafcat so that the relationship of the straight line (1) shown in FIG. 4 is satisfied. Note that when the inclination of the catalyst characteristic correction exceeds the inclination of the straight line (1), the upstream air-fuel ratio becomes too rich, and the amount of HC and NOx emissions increases rapidly. The straight line (1) shown in FIG. 4 has a slope immediately before the amount of discharge increases rapidly.

  As described above, the system of the present embodiment calculates the catalyst characteristic correction amount evafcat by referring to the map shown in FIG. 3 (see step 110 above). Here, the map shown in FIG. 3 shows that in the region where the intake air amount Ga exceeds 15 (g / sec), the catalyst characteristic correction amount evafcat is the intake air amount Ga with the same slope of “1.6” as the straight line (1). It is stipulated to increase or decrease. That is, in the region where the intake air amount Ga exceeds 15 (g / sec), the upstream air-fuel ratio is determined to be appropriately richer than the actual stoichiometric air-fuel ratio of the upstream catalyst 12.

  15 (g / sec) described above is a boundary value that divides a region in which the enrichment of the upstream air-fuel ratio leads to an improvement in emission characteristics and a region in which the improvement effect cannot be obtained so much. Therefore, according to the map shown in FIG. 3, in the region where the intake air amount Ga is large to a certain extent that enriching the upstream air-fuel ratio is effective in improving the emission characteristics, the catalyst characteristic correction amount evafcat is increased by a large slope (1 .6). For this reason, according to the system of the present embodiment, it is possible to stably realize good emission characteristics during high load operation in which the intake air amount Ga is large.

(Explanation of sub-feedback learning)
Next, the contents of sub feedback learning executed in the present embodiment will be described. As described above, the system of this embodiment calculates the sub feedback correction amount evafsfb so that the output of the sub O ₂ sensor 18 becomes a stoichiometric output (see step 104 above), and also performs sub feedback correction at a predetermined timing. The average value of the quantity evafsfb is transferred to the sub-feedback learning quantity evafsfbg (see step 108 above).

  The sub-feedback correction amount evafsfb is a value corresponding to the difference between the upstream air-fuel ratio for making the downstream air-fuel ratio stoichiometric and the actual upstream air-fuel ratio. The average value of the sub-feedback correction amount evafsfb corresponds to a permanent deviation inherent in the air-fuel ratio control (deviation caused by individual differences or changes with time of the internal combustion engine 10).

  If the average value of the sub-feedback correction amount evafsfb is transferred to the sub-feedback learning amount evafsfbg, the permanent deviation inherent in the system can be offset by the sub-feedback learning amount evafsfbg. In this case, the role of the sub-feedback correction amount evafsfb is only to absorb temporary deviations caused by various causes. If the sub-feedback correction amount evafsfb is also responsible for absorption of constant deviation in the system, it takes a long time to converge. On the other hand, if the role is only to absorb temporary deviation, the time required for the convergence can be sufficiently shortened. For this reason, transferring part of the sub-feedback correction amount evafsfb to the sub-feedback learning amount evafsfbg is extremely useful for shortening the control convergence time.

  By the way, as described above, the system of the present embodiment adds the catalyst characteristic correction amount evafcat to the output evafbse of the main air-fuel ratio sensor 16 when calculating the corrected A / F output evabyf. For this reason, in the region where the intake air amount Ga exceeds 15 (g / sec), the downstream air-fuel ratio is enriched due to the influence of the catalyst characteristic correction amount evafcat.

In the system of the present embodiment, when the downstream air-fuel ratio is enriched, the output Voxs of the sub O ₂ sensor 18 becomes a rich output, and the sub feedback correction amount evafsfb is updated so that the output voxs becomes a stoichiometric output. That is, in the system of the present embodiment, in the region where the intake air amount Ga exceeds 15 (g / sec), control for enriching the downstream air-fuel ratio by the action of the catalyst characteristic correction amount evafcat and the enrichment are performed. There arises a situation in which sub-feedback control for eliminating the downstream air-fuel ratio to be stoichiometric is executed simultaneously.

  If the target of the air-fuel ratio control is to make the downstream air-fuel ratio stoichiometric, it can be recognized that the air-fuel ratio deviation that appears in the downstream air-fuel ratio is a steady deviation inherent in the air-fuel ratio control itself. For this reason, when the above target is applied, it is meaningful to transfer the average value of the sub feedback correction amount evafsfb to the sub feedback learning amount evafsfbg as being caused by a constant error.

  However, under a situation where the downstream air-fuel ratio is intentionally enriched, and under a situation where the degree of intentional enrichment varies depending on the intake air amount Ga, the sub feedback correction amount evafsfb There is no close correlation between the average value and the constant deviation of the air-fuel ratio control itself. If learning of the sub-feedback learning amount evafsfbg proceeds under such circumstances, the value deviates from the value for absorbing the constant deviation occurring in the system, and the accuracy of air-fuel ratio control is on the contrary. Cause it to worsen.

  Therefore, in the present embodiment, in the region where the intake air amount Ga is less than 15 (g / sec), the relationship between the catalyst characteristic correction amount evafcat and the intake air amount Ga is on the straight line y = 0.8x, that is, The map of the catalyst characteristic correction amount evafcat is set so that the upstream air-fuel ratio becomes the actual stoichiometric air-fuel ratio of the upstream catalyst 12 (see FIG. 3). In addition, learning of the sub-feedback learning amount evafsfbg is permitted only when Ga ≦ 15 (g / sec) is established (see step 106 above).

  Since the map shown in FIG. 3 is set as described above, in the region where the intake air amount Ga is less than 15 (g / sec), the catalyst characteristic correction amount evafcat has the relationship of the straight line (2) shown in FIG. Set to a value that satisfies. According to the catalyst characteristic correction amount evafcat that satisfies the relationship of the straight line (2), the upstream air-fuel ratio can be appropriately raised or lowered as the catalyst window rises or falls according to the intake air amount Ga. That is, according to the catalyst characteristic correction amount evafsfb that satisfies the relationship of the straight line (2), it is possible to prevent the downstream air-fuel ratio from becoming rich or lean due to the dependence of the catalyst window on the intake air amount Ga. .

Under such circumstances, the sub O ₂ sensor 18 becomes rich or lean mainly due to a constant deviation occurring in the system. In this case, the average value of the sub feedback control amount evafsfb is a value having a close correlation with the constant deviation. In the system of the present embodiment, the sub feedback learning amount evafsfbg is allowed to be updated only when the average value of the sub feedback control amount evafsfb has such characteristics. Therefore, according to the system of the present embodiment, it is possible to accurately learn a value for offsetting a constant deviation occurring in the system as the sub-feedback learning amount evafsfbg.

FIG. 6 is a timing chart for explaining an effect obtained by limiting the learning region of the sub feedback learning amount evafsfbg to a region where the intake air amount Ga is 15 (g / sec) or less. More specifically, FIG. 6 (D) shows the vehicle speed pattern during the test run, and FIGS. 6 (A) and 6 (B) show the sub feedback correction amount evafsfb generated for the vehicle speed pattern. , And the waveform of the output voxs of the sub O ₂ sensor 18. The two waveforms shown in FIG. 6C are a waveform of the sub-feedback learning amount evafsfbg when the learning region is limited (upper), and a waveform of the sub-feedback learning amount evafsfbg when learning is permitted in the entire region ( Down).

  As shown in FIG. 6C, when learning is performed in the entire area, the sub feedback learning amount evafsfbg shows a large change. On the other hand, when the learning region is limited to the low Ga region, the sub feedback learning amount evafsfbg maintains a stable value. As is clear from these waveforms, if the learning region is limited to the low Ga region, the sub feedback learning amount evafsfbg can be prevented from being updated to an inappropriate value without being affected by the running state of the vehicle. . For this reason, according to the system of the present embodiment, it is possible to realize air-fuel ratio control with excellent convergence.

In the first embodiment described above, the upstream catalyst 12 is the “catalyst” in the first invention, the main air-fuel ratio sensor 16 is the “upstream exhaust gas sensor” in the first invention, and the sub O ₂ sensor. 18 corresponds to the “downstream exhaust gas sensor” in the first invention, the air flow meter 22 corresponds to the “intake air amount detection means” in the first invention, and the ECU 20 performs the routine shown in FIG. By executing this, the “fuel injection amount control means” according to the first aspect of the present invention is realized. Here, FIG. 4 shows the catalyst characteristic correction amount evafcat that satisfies the relationship y = 1.6x-12 shown in FIG. 3 with 15 (g / sec) being the “predetermined value” in the first invention. The value converted into the air-fuel ratio in accordance with the rules is the “predetermined air-fuel ratio” in the first invention, and the characteristic that minimizes the amount of HC and NOx generated as a whole is the “desired emission characteristic” in the first invention. , Respectively.

  In the first embodiment described above, the ECU 20 executes the process of step 100, so that the “means for calculating the feedback correction basic quantity” in the second invention executes the process of step 104. Accordingly, the “means for calculating the sub-feedback correction amount” in the second invention performs the processing of step 110, so that the “catalyst characteristic correction amount calculation means” in the second invention becomes the step of step 112. By executing the processing, the “corrected correction amount calculating means” in the second invention performs the processing of steps 114 and 116, and “the means for calculating the fuel injection amount” in the second invention is executed. , Each has been realized.

  In the first embodiment described above, the ECU 20 executes the process of step 108 so that the “means for calculating the sub-feedback learning amount” in the third invention executes the process of step 112. As a result, the “corrected correction amount calculating means” in the third invention realizes the “learning area limiting means” in the third invention by executing the processing of step 106.

Embodiment 2. FIG.
[Features of Embodiment 2]
Next, a second embodiment of the present invention will be described with reference to FIGS. The system of the present embodiment can be realized by causing the ECU 20 to execute a routine shown in FIG. 7 described later using the hardware shown in FIG.

In the first embodiment described above, the purpose of the sub-feedback control is to set the downstream air-fuel ratio to the actual stoichiometric air-fuel ratio, that is, to generate the stoichiometric output in the sub-O ₂ sensor 18. Therefore, in the system of the first embodiment, when the downstream air-fuel ratio is intentionally enriched using the catalyst characteristic correction amount evafcat, the sub feedback correction amount evafsfb is updated so as to cancel the enrichment. The phenomenon occurs. In the first embodiment, the learning region is limited to the low Ga region in order to avoid inappropriate updating of the sub feedback learning amount evafsfbg in the process of updating the sub feedback correction amount evafsfb as described above. I am going to do that.

  On the other hand, in the present embodiment, when the upstream air-fuel ratio is intentionally enriched, the sub feedback control is executed so that the intentional rich amount is not canceled. That is, the influence of the enrichment of the downstream air-fuel ratio accompanying the intentional enrichment of the upstream air-fuel ratio is excluded from the sub-feedback control amount evafsfb, and the sub-feedback control is executed.

  According to such control, even when the upstream air-fuel ratio is intentionally enriched, the sub-feedback correction amount evafsfb is always not affected by the intentional enrichment, and is a deviation inherent in the system. And is kept at a value closely correlated. Therefore, according to the system of the present embodiment, learning of the sub-feedback learning amount evafsfbg can be permitted over the entire area while performing control to intentionally enrich the upstream air-fuel ratio in order to stabilize the emission characteristics. it can.

[Specific Processing in Second Embodiment]
FIG. 7 is a flowchart of a routine executed in the present embodiment in order to realize the above function. In the routine shown in FIG. 7, first, the catalyst characteristic correction amount evafcat2 is calculated (step 120). The catalyst characteristic correction amount evafcat2 calculated here is a value satisfying the relationship of the straight line (2) shown in FIGS. 3 and 4, that is, the relationship of y = 0.8x with respect to the intake air amount Ga. In the present embodiment, the ECU 20 stores a map of the catalyst characteristic correction amount evafcat2 set so as to satisfy this relationship. In step 120, the catalyst characteristic correction amount evafcat2 corresponding to the current Ga is calculated by referring to the map.

  In the routine shown in FIG. 7, next, the rich correction amount evafrich is calculated (step 122). FIG. 8 shows a map of the rich correction amount evafrich stored in the ECU 20. This map corresponds to the difference between the straight line (1) and the straight line (2) shown in FIG. 4. If the rich correction amount evafrich is the y value and the intake air amount Ga is the x value, y = (1.6−0.8). ) Satisfies the relationship of x. In step 122, the rich correction amount evafrich is calculated with reference to the map shown in FIG.

  Next, according to the routine shown in FIG. 7, the output evafbse of the main air-fuel ratio sensor 16 is detected (step 124), and further, the processing related to the stoichiometric point learning amount evafofs is executed (step 126). Since these processes are the same as the processes of steps 100 and 102 shown in FIG. 2, detailed description thereof is omitted here.

  Next, the calculation process of the sub feedback correction amount evafsfb and the learning process of the sub feedback learning amount evafsfbg are sequentially executed (steps 128 and 130). Specifically, these processes are executed according to the procedure shown in FIG. The processing executed here will be described in detail later with reference to FIG. 9 and FIG.

In the routine shown in FIG. 7, the corrected A / F output evabyf is then calculated according to the following arithmetic expression (step 132).
evabyf = evafbse + evafofs + evafsfb + evafsfbg + evafcat2 + evafrich
... (2)

  The above equation (2) is the same as the equation (1) used in the first embodiment except that “evafcat” is replaced by “evafcat2 + evafrich”. “Evafcat2 + evafrich” is a value that satisfies the relationship of the straight line (1) shown in FIG. For this reason, according to the corrected A / F output evabyf calculated by the above equation (2), when the intake air amount Ga changes, the upstream air-fuel ratio is determined without waiting for the sub-feedback correction amount evafsfb to follow. It can be changed along the straight line (1) shown in FIG. 4 (however, in this case, the vertical axis in FIG. 4 is regarded as the air-fuel ratio).

  According to the routine shown in FIG. 7, the calculation process of the control target air-fuel ratio eabyfref and the main feedback process are sequentially performed thereafter (steps 134 and 136). These processes are executed in the same manner as the processes of steps 114 and 116 shown in FIG.

(Calculation of sub feedback correction amount and sub feedback learning amount)
Next, the sub feedback correction amount evafsfb calculation process and the sub feedback learning amount evafsfbg calculation process in the present embodiment will be described in detail with reference to FIGS. 9 and 10.

In the routine shown in FIG. 9, first, the output deviation dlvoxs of the sub O ₂ sensor is calculated (step 140). Output deviation Dlvoxs is calculated as dlvoxs = voxsref-voxs by subtracting the output Voxs of the sub O ₂ sensor 18 from the stoichiometric output Voxsref sub O ₂ sensor 18 (see FIG. 10).

Next, an integration process of the output deviation dlvoxs is executed (step 142). Specifically, the deviation integrated value dlvoxssum is calculated according to the following arithmetic expression.
dlvoxssum = dlvoxssum (i-1) + dlvoxs + (evafrich / GainI) (3)

  However, in the above equation (3), dlvoxssum (i-1) is the deviation integral value calculated during the previous processing cycle. Gain I is a gain (hereinafter referred to as “integral term gain”) multiplied by the deviation integral value dlvoxssum when calculating the integral correction amount evafsfbi which is a component of the sub feedback correction amount evafsfb.

  According to Equation (3) above, the deviation integrated value dlvoxssum (i-1) calculated during the previous processing cycle is divided by the integral gain GainI and the currently detected output deviation dlvoxs and the rich correction amount evafrich used this time. Is added. As described, the system of the present embodiment calculates the corrected A / F output evabyf according to the above equation (2). Here, the catalyst characteristic correction value evafcat2 added to the right side of the equation (2) enriches the upstream air-fuel ratio so that the upstream air-fuel ratio becomes the actual stoichiometric air-fuel ratio regardless of the intake air amount Ga. Is the correction amount. The rich correction amount evafrich added to the right side of the equation (2) is a correction amount for making the upstream air-fuel ratio richer than the actual stoichiometric air-fuel ratio. Therefore, in the system of the present embodiment, the upstream air-fuel ratio is intentionally richer than the actual stoichiometric air-fuel ratio by the rich correction amount evafrich.

In a situation where the upstream air-fuel ratio is intentionally enriched by the rich correction amount evafrich, the effect of the enrichment also appears in the downstream air-fuel ratio. For this reason, the output voxs of the sub O ₂ sensor 18 becomes a value (large value) biased richly by the influence of the rich correction amount evafrich, compared to that caused by a system shift. As a result, the output deviation dlvoxs is smaller than the original value by the influence of the rich correction amount evafrich.

In the above equation (3), the excess due to the rich correction amount evafrich is eliminated by adding (evafrich / GainI) to the output deviation dlvoxs. In other words, in the above equation (3), the term “dlvoxs + (evafrich / GainI)” compensates for the subtle caused by the rich correction amount evafrich, and thereby the output deviation amount of the sub O ₂ sensor 18 caused by the deviation of the system. Is a value that accurately represents. Therefore, according to the above equation (3), the output deviation amount of the sub O ₂ sensor 18 caused by the deviation of the system can be integrated, and the integrated value can be calculated as the deviation integrated value dlvoxssum.

In the routine shown in FIG. 9, the change amount dvoxs generated in the output voxs of the sub O ₂ sensor 18 from the previous cycle to the current cycle is calculated (step 144). Specifically, the output change amount dvoxs is calculated by subtracting the current output voxs (i) from the previous output voxs (i−1) as shown in the following equation. The output change amount dvoxs calculated in this way has significance as a differential value of the output voxs of the sub O ₂ sensor 18.
dvoxs = voxs (i−1) −voxs (i) (4)

Next, a proportional correction amount evafsfbp, which is a first component of the sub feedback correction amount evafsfb, is calculated (step 146). Specifically, the proportional correction amount evafsfbp is calculated according to the following equation. However, GainP included in the following equation is a proportional gain used for calculating the proportional correction amount evafsfbp.
evafsfbp = dlvoxs * GainP + evafrich (5)

The proportional correction amount evafsfbp is a correction amount for feeding back the difference between the actual output voxs of the sub O ₂ sensor 18 and the target value to the content of the air-fuel ratio control. In the present embodiment, as described above, the downstream air-fuel ratio is enriched due to the influence of the rich correction amount evafrich. As a result, the output voxs of the sub O ₂ sensor 18 has an excessive value due to the influence of the rich correction amount evafrich, and the output deviation dlvoxs is only the influence of the rich correction amount evafrich than the value indicating the deviation of the system. The value is too small.

  According to the above equation (5), by adding the rich correction amount evafrich to dlvoxs * GainP, it is possible to appropriately compensate for the shortage of the output deviation dlvoxs. Therefore, according to the above equation (5), it is possible to calculate the rich correction amount evafrich that is closely correlated with the deviation amount of the air-fuel ratio due to the deviation of the system.

In the routine shown in FIG. 9, next, the integral correction amount evafsfbi, which is the second component of the sub readback correction amount evafsfb, is calculated (step 148). Here, specifically, the integral correction amount evafsfbi is calculated according to the following equation.
evafsfbi = dlvoxssum * GainI (6)

  In step 142, as described above, the deviation integral value dlvoxssum excluding the influence of the rich correction amount evafrich can be calculated. Therefore, according to the above equation (6), the integral correction amount evafsfbi suitable for correcting the air-fuel ratio bias caused by the system deviation is calculated without being affected by the richness by the rich correction amount evafrich. be able to.

Next, a differential correction amount evafsfbd, which is a third component of the sub feedback correction amount evafsfb, is calculated (step 150). Specifically, the differential correction amount evafsfbd is calculated as follows by multiplying the output change amount dvoxs of the sub O ₂ sensor 18 by the differential gain GainD.
evafsfbd = dvoxs * GainD (7)

Note that the enrichment of the upstream air-fuel ratio due to the rich correction amount evafrich does not affect the changing speed of the downstream air-fuel ratio. That is, the enrichment caused by the rich correction amount evafrich does not significantly affect the differential value of the output voxs of the sub O ₂ sensor 18. Therefore, the differential correction amount evafsfbd can be calculated with high accuracy without performing processing for eliminating the influence of the rich correction amount evafrich.

When the above processing is completed, the sub feedback correction amount evafsfb is calculated (step 152). The sub feedback correction amount evafsfb is calculated as follows by adding the proportional correction amount evafsfbp, the integral correction amount evafsfbi, and the differential correction amount evafsfbd.
evafsfb = evafsfbp + evafsfbi + evafsfbd (8)

  According to the above processing, the sub feedback correction amount evafsfb can be calculated as a value from which the influence of intentional enrichment due to the rich correction amount evafrich is eliminated. If the sub-feedback correction amount evafsfb is a value that is not affected by the intentional enrichment, the sub-feedback correction amount evafsfb cancels the enrichment even if the sub-feedback control is executed simultaneously with the intentional enrichment. Will not be updated.

  That is, under such a situation, the sub-feedback control does not function to cancel intentional enrichment. Therefore, according to the system of the present embodiment, the upstream air-fuel ratio is controlled to the originally intended rich air-fuel ratio with extremely high accuracy by simultaneously executing the enrichment control using the rich correction amount evafrich and the sub-feedback control. Is possible.

  Further, the value of the sub-feedback correction amount evafsfb from which the influence of the intentional enrichment has been eliminated shows a close correlation with the magnitude of the deviation of the downstream air-fuel ratio caused by the system deviation. For this reason, in the system of the present embodiment, while performing intentional enrichment using the rich correction amount evafrich, the steady-state value of the sub feedback correction amount evafsfb is always expressed as a sub feedback learning amount that represents a system deviation. Can be imported into evafsfbg.

  In the routine shown in FIG. 9, the process of step 128 shown in FIG. 7, that is, the calculation process of the sub feedback correction amount evafsfb is executed by the above procedure. When these processes are completed, the process of Step 154 is then executed to perform the process of Step 130 shown in FIG. 7, that is, the process of calculating the sub-feedback learning amount evafsfbg.

Here, first, the sub feedback learning smoothed value evafsfbsm is calculated according to the following equation (step 154).
evafsfbsm = evafsfbsm (i−1) + {evafsfbp−evafsfbsm (i−1)} / n (9)

  The above equation (9) is an arithmetic expression for reflecting the difference between the current proportional correction amount evafsfbp and the previous sub-feedback learning smoothed value to the sub-feedback learning smoothed value with a gain of 1 / n. . That is, this arithmetic expression (9) is an expression for setting the value obtained by dividing the proportional correction amount evafsfbp by the gain of 1 / n as the sub-feedback learning smoothed value evafsfbsm.

Of the elements constituting the sub feedback correction amount evafsfb, the differential correction amount evafsfbd is a value that is greatly influenced by the output characteristics of the sub O ₂ sensor 18 and is a value that is unrelated to the steady-state deviation of the system. For this reason, in the present embodiment, the differential correction amount evafsfbd is not included in the sub-feedback learning amount evafsfbg.

  Further, the integral correction amount evafsfbi, which is another component of the sub feedback correction amount evafsfb, is likely to include a larger error than the proportional correction amount evafsfbp. For this reason, the integral correction amount evafsfbi is not included in the sub-feedback learning amount evafsfbg and is therefore not reflected in the sub-feedback learning smoothed value evafsfbsm.

However, since the integral correction amount evafsfbi is a value representing the steady-state deviation of the system, it may be taken into the sub feedback learning amount evafsfbg together with the proportional correction amount evafsfbp or independently. In the above step 154, the former content can be realized by calculating the sub feedback learning simulated value by the following equation (10), and the latter content can be realized by calculating by the following equation (11). .
evafsfbsm = evafsfbsm (i−1) + {evafsfbp + evafsfbi−evafsfbsm (i−1)} / n
(10)
evafsfbsm = evafsfbsm (i−1) + {evafsfbi−evafsfbsm (i−1)} / n (11)

  In the routine shown in FIG. 9, it is then determined whether or not the learning timing for the sub feedback learning amount evafsfbg has arrived (step 156). As a result, if it is determined that the timing has not yet arrived, the current cycle is terminated as it is. On the other hand, if the arrival of the learning timing is recognized, the update amount edvafsfbg of the sub feedback learning amount evafsfbg is calculated (step 158).

Specifically, the update amount edvafsfbg is calculated according to the following equation (12). According to this equation (12), 1 / m of the sub feedback learning smoothed value evafsfbsm can be set as the update amount edvafsfbg.
edvafsfbg = evafsfbsm / m (12)

Next, update processing of the sub feedback learning value evafsfbg is executed (step 160). Here, by adding the update amount edvafsfbg to the learned value evafsfb (i−1) in the previous cycle, the update process is performed as in the following equation.
evafsfbg = evafsfbg (i−1) + edvafsfbg (13)

Finally, learning value update post-processing is executed in preparation for the subsequent processing (step 162). Here, in response to the addition of the update amount evafsfbg to the update amount sub-feedback learning value evafsfbg, (a) the sub-feedback correction amount evafsfb, (b) the deviation integral value dlvoxssum, and (c) the sub-feedback learning The correct value evafsfbsm is corrected based on the update amount edvafsfbg. Specifically, these corrections are executed according to the following arithmetic expression.
evafsfb = evafsfb−edvafsfbg (14)
dlvoxssum = dlvoxssum− (edvafsfbg / GainI) (15)
evafsfbsm = evafsfbsm−edvafsfbg (16)

  As described above, according to the routines shown in FIGS. 7 and 9, the control using the rich correction amount evafrich consciously enriches the upstream air-fuel ratio and eliminates the influence of the conscious enrichment. The feedback correction amount evafsfb can be calculated. Then, the sub-feedback learning value evafsfbg can be properly learned by shifting the steady value included in the sub-feedback correction amount evafsfb. For this reason, according to the system of this embodiment, it is possible to always achieve excellent emission characteristics without being affected by the amount of intake air amount Ga.

  In the second embodiment described above, the ECU 20 executes the process of step 124, so that the “means for calculating the feedback correction basic amount” in the fourth invention executes the process of step 120. As a result, the “characteristic correction amount calculating means” in the fourth aspect of the invention executes the processing of step 122, whereby the “rich correction amount calculating means” of the fourth aspect of the invention is the processing of step 128, that is, By executing the processing of steps 140 to 152, the “sub-feedback correction amount calculating means” in the fourth aspect of the invention executes the processing of step 132, and “calculates the correction amount after correction” of the fourth aspect of the invention. The means "executes the processes of steps 134 and 136 described above, whereby" means " Fuel injection quantity means for calculating a "are realized respectively.

Embodiment 3 FIG.
[Features of Embodiment 3]
Next, Embodiment 3 of the present invention will be described with reference to FIG. 11 and FIG. The system of the present embodiment can be realized by causing the ECU 20 to further execute a routine shown in FIG. 11 described later in the system of the second embodiment.

  In the system of the first embodiment described above, the upstream air-fuel ratio is consciously enriched in the high Ga region by adding the catalyst characteristic correction amount evafcat to the output evafbse of the main air-fuel ratio sensor 12. In the second embodiment described above, the upstream air-fuel ratio is consciously enriched by adding the catalyst characteristic correction amount evafcat2 and the rich correction amount evafrich to the output of the main air-fuel ratio sensor 12.

  In these systems, if the catalyst characteristic correction amounts evafcat, evafcat2 and the rich correction amount evafrich are excessive, the upstream air-fuel ratio becomes unduly rich. More specifically, in these systems, the upstream air-fuel ratio is excessively enriched, which may cause a situation where the downstream air-fuel ratio is clearly unduly enriched. Therefore, the system of the present embodiment places a limit on conscious enrichment of the upstream air-fuel ratio so that the downstream air-fuel ratio does not unduly deviate from stoichiometry.

[Specific Processing in Embodiment 3]
FIG. 11 is a flowchart of a routine executed by the ECU 20 in the present embodiment in order to realize the above function. Specifically, this routine is executed as the process of step 122 shown in FIG. 7, that is, as a process for calculating the rich correction amount evafrich.

  In the routine shown in FIG. 11, first, the base value of the rich correction amount evafrich is calculated based on the intake air amount Ga (step 170). Here, the value read with reference to the map shown in FIG. 8 is set as the base value of the rich correction amount evafrich.

Next, it is determined whether or not the output voxs of the sub O ₂ sensor 18 detected in the current cycle exceeds the maximum output voxsmax detected in the past (step 172). As a result, when the establishment of voxs> voxsmax is recognized, the current voxs is rewritten to the maximum output voxsmax (step 174). On the other hand, if the above determination is negative, the process of step 174 is jumped.

Next, the upper limit value of the proportional correction amount evafsfbp, which is one element of the sub feedback correction amount evafsfb, is calculated (step 176). Here, specifically, the upper limit value is calculated according to the following equation.
Upper limit = (voxsmax−voxsref) * 0.8 * GainP (17)

FIG. 12 is a diagram for explaining the physical meaning of the equation (17). As shown in FIG. 12, the maximum output voxsmax is a convergence value on the upper limit side of the output voxs of the sub O ₂ sensor 18. Therefore, the output voxs being the maximum output voxsmax means that the downstream air-fuel ratio is clearly a rich air-fuel ratio. The fact that the downstream air-fuel ratio becomes a clear rich air-fuel ratio is not a preferable state for obtaining good emission characteristics. Therefore, even when the upstream air-fuel ratio is consciously enriched, the enrichment should be performed within a range in which the downstream air-fuel ratio does not reach an obvious rich air-fuel ratio.

The term “(voxsmax−voxsref) * 0.8” in the equation (17) is allowed for the output deviation dlvoxs of the sub O ₂ sensor 18 when the upstream air-fuel ratio is consciously enriched. The upper limit value is a value set from the above viewpoint. When the upper limit value allowed for the output deviation dlvoxs is “(voxsmax−voxsref) * 0.8”, the upper limit value allowed for the proportional correction amount evafsfbp is expressed by the above equation (17). It will be a thing.

In the routine shown in FIG. 11, next, the process of updating the minimum output voxsmin of the sub O ₂ sensor 18 (steps 178 and 180) and the process of calculating the lower limit value max of the proportional correction amount evafsfbp according to the following equation (step 182). ) Are performed sequentially.
Lower limit value = (voxsmin−voxsref) * 0.8 * GainP (18)
These processes are executed for the same purpose in the same procedure as the processes in steps 172 to 176 described above.

  When the above processing ends, it is next determined whether or not the base value of the rich correction amount evafrich calculated in step 170 is larger than the upper limit value of the proportional correction amount evafsfbp (step 184).

  When the rich correction amount evafrich is larger than the upper limit value, it is requested that enrichment that may not be corrected by the proportional correction amount evafsfbp is requested, and it can be determined that the request is excessive. In this case, the rich correction amount evafrich is replaced with the above upper limit value in order to make the request for enrichment reasonable (step 186). On the other hand, if evafrich> the upper limit value is not satisfied, the processing of step 186 is jumped assuming that the enrichment request is appropriate.

  From the same viewpoint, it is next determined whether or not the enrichment correction amount evafsfb is below the lower limit value of the proportional correction amount evafsfbp (step 188). If evafsfb <lower limit value is satisfied, processing for replacing the enrichment correction amount evafrich with the above lower limit value is performed (step 190).

  According to the above processing, no matter what enrichment correction amount evafrich (base value) is set based on the intake air amount Ga, the shift amount consciously given to the upstream air-fuel ratio is the same as the downstream air-fuel ratio. It will be limited to the range that can be maintained in the vicinity of stoichi. For this reason, according to the system of the present embodiment, the exhaust gas greatly deviating from the stoichiometry from the upstream catalyst 12 flows out due to the execution of the enrichment control using the rich correction amount evafrich. Can be effectively prevented.

  In the third embodiment, the ECU 20 executes the process of step 120 shown in FIG. 7 so that the “characteristic correction amount calculating means” in the fifth aspect of the invention executes the process of step 170. Thus, the “rich correction amount calculating means” in the fifth invention adds the evafcat2 and evafrich in step 132 shown in FIG. 7 to thereby add the “means for calculating the catalyst characteristic correction amount” in the fifth invention. By executing the processing of steps 172 and 174, the “means for detecting the output upper limit value” in the fifth aspect of the invention executes the processing of step 176 of the “rich correction amount” in the fifth aspect of the invention. The “means for calculating the upper limit guard value” executes the processing of step 186 described above, thereby increasing the “rich correction amount upper limit” in the fifth aspect of the invention. A means for replacing with a limit guard value "is realized.

  In the third embodiment described above, the ECU 20 executes the process of step 120 shown in FIG. 7 so that the “characteristic correction amount calculating means” in the sixth aspect of the invention executes the process of step 170. Accordingly, the “rich correction amount calculating means” in the sixth invention adds the evafcat2 and evafrich in step 132 shown in FIG. 7 to thereby add the “means for calculating the catalyst characteristic correction amount” in the sixth invention. By executing the processing of steps 178 and 180, the “means for detecting the output lower limit value” in the sixth aspect of the invention executes the processing of step 182 so as to execute the “rich correction amount in the sixth aspect of the invention. The “means for calculating the lower limit guard value” executes the process of step 190 described above, thereby performing the “rich correction amount” Means for replacing with the lower limit guard value are realized.

It is a figure for demonstrating the structure of the air fuel ratio control apparatus of Embodiment 1 of this invention. It is a flowchart of the routine performed in Embodiment 1 of this invention. 6 is a map that defines a relationship between an intake air amount Ga and a catalyst characteristic correction amount evafcat. It is a figure for demonstrating four straight lines (1)-(4) considered in making the map of the catalyst characteristic correction amount evafcat. It is the result of the experiment which confirmed the relationship between the inclination of catalyst characteristic correction | amendment and the discharge | emission amount of HC and NOx. FIG. 10 is a timing chart for explaining the effect of limiting the learning region of the sub feedback learning amount evafsfbg to a region where the intake air amount Ga is 15 (g / sec) or less. It is a flowchart of the routine performed in Embodiment 2 of this invention. 7 is a map of a rich correction amount evafrich used in Embodiment 2 of the present invention. FIG. 8 is a flowchart of a routine showing details of steps 128 and 130 shown in FIG. 7. FIG. It is a figure for demonstrating the content of the parameter used for calculation of the sub feedback correction amount evafsfb. It is a flowchart of the routine performed in Embodiment 3 of the present invention. It is a figure for demonstrating the physical meaning of the upper limit and lower limit of the proportional correction amount evafsfbp.

Explanation of symbols

10 internal combustion engine 12 upstream catalyst 16 main air-fuel ratio sensor 18 sub O ₂ sensor 20 ECU (Electronic Control Unit)
Ga intake air volume
evafbse Main air-fuel ratio sensor output
evafsfb Sub feedback correction amount
evafsfbg Sub feedback learning amount
evafcat ； evafcat2 Catalyst characteristic correction amount
A / F output after evabyf correction
evafrich Rich correction amount
voxs Sub O ₂ sensor output
dlvoxs output deviation
voxsref stoiki output
dlvoxssum Deviation integral value
Evafsfbp proportional correction amount
evafsfbi integral correction amount

Claims (6)

A catalyst disposed in an exhaust passage of the internal combustion engine;
An upstream exhaust gas sensor disposed upstream of the catalyst;
A downstream side exhaust gas sensor disposed downstream of the catalyst;
Intake air amount detection means for detecting the intake air amount of the internal combustion engine;
Fuel injection amount control means for controlling the fuel injection amount based on the output of the upstream side exhaust gas sensor and the output of the downstream side exhaust gas sensor;
The fuel injection amount control means adjusts the fuel injection amount so that the air-fuel ratio upstream of the catalyst becomes a predetermined air-fuel ratio richer than the actual stoichiometric air-fuel ratio of the catalyst in a region where the intake air amount exceeds a predetermined value. Control,
An air-fuel ratio control apparatus for an internal combustion engine, wherein the predetermined air-fuel ratio is a value determined in relation to an intake air amount as a means for realizing desired emission characteristics. The fuel injection amount control means includes
Means for calculating a feedback correction basic amount for setting the air-fuel ratio upstream of the catalyst to the target air-fuel ratio based on the output of the upstream side exhaust gas sensor;
Means for calculating a sub feedback correction amount for matching them based on the difference between the output of the downstream side exhaust gas sensor and the stoichiometric output;
Catalyst characteristic correction amount calculating means for setting a catalyst characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the predetermined air-fuel ratio based on the intake air amount;
A corrected correction amount calculating means for calculating a corrected feedback correction amount by correcting the feedback correction basic amount based on the sub feedback correction amount and the catalyst characteristic correction amount;
Means for calculating the fuel injection amount based on the post-correction feedback correction amount;
The air-fuel ratio control apparatus for an internal combustion engine according to claim 1, further comprising: The fuel injection amount control means includes
In the region where the intake air amount is less than or equal to the predetermined value, the fuel injection amount is controlled so that the air-fuel ratio upstream of the catalyst becomes the actual stoichiometric air-fuel ratio of the catalyst, and
Means for calculating a sub feedback learning amount by taking a steady value included in the sub feedback correction amount;
A corrected correction amount calculating means for calculating the feedback correction basic amount based on the sub feedback learning amount in addition to the sub feedback correction amount and the catalyst characteristic correction amount;
The air-fuel ratio control apparatus for an internal combustion engine according to claim 2, further comprising learning region limiting means for permitting the update of the sub feedback learning amount only when the intake air amount is equal to or less than the predetermined value. The fuel injection amount control means includes
Means for calculating a feedback correction basic amount for setting the air-fuel ratio upstream of the catalyst to the target air-fuel ratio based on the output of the upstream side exhaust gas sensor;
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Sub feedback correction amount calculation means for calculating a value obtained by eliminating the influence of the rich correction amount as a sub feedback correction amount from a correction amount for matching the output of the downstream exhaust gas sensor and the stoichiometric output;
A corrected correction amount calculating means for calculating a corrected feedback correction amount by correcting the feedback correction basic amount based on the characteristic correction amount, the rich correction amount, and the sub feedback correction amount;
Means for calculating the fuel injection amount based on the post-correction feedback correction amount;
The air-fuel ratio control apparatus for an internal combustion engine according to claim 1, further comprising: The catalyst characteristic correction amount calculating means includes:
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Means for calculating the catalyst characteristic correction amount by adding the characteristic correction amount and the rich correction amount;
Means for detecting an output upper limit value of the downstream side exhaust gas sensor;
Means for calculating a rich correction amount upper limit guard value smaller than the output upper limit value based on the output upper limit value;
Means for replacing the rich correction amount with the rich correction amount upper limit guard value when the rich correction amount is larger than the rich correction amount upper limit guard value;
The air-fuel ratio control apparatus for an internal combustion engine according to any one of claims 2 to 4, further comprising: The catalyst characteristic correction amount calculating means includes:
Characteristic correction amount calculation means for setting a characteristic correction amount for setting the air-fuel ratio upstream of the catalyst to the actual stoichiometric air-fuel ratio of the catalyst based on the intake air amount;
Rich correction amount calculation means for setting a rich correction amount based on the intake air amount to enrich the air-fuel ratio upstream of the catalyst by the difference between the actual theoretical air-fuel ratio of the catalyst and the predetermined air-fuel ratio;
Means for calculating the catalyst characteristic correction amount by adding the characteristic correction amount and the rich correction amount;
Means for detecting an output lower limit value of the downstream side exhaust gas sensor;
Means for calculating a rich correction amount lower limit guard value larger than the output upper limit value based on the output lower limit value;
Means for replacing the rich correction amount with the rich correction amount lower limit guard value when the rich correction amount is larger than the rich correction amount guard value;
The air-fuel ratio control apparatus for an internal combustion engine according to any one of claims 2 to 5, further comprising:

Images