Control device for internal combustion engine

Abstract

To provide a control device of an internal combustion engine which can effectively prevent the deterioration of emission.SOLUTION: A control device 34 calculates a detection absorption / discharge amount OSR# on the basis of an AF detection value O# being an actual output value of a front-side AF sensor 28, calculates a guard absorption / discharge amount OSRg on the basis of an AF ideal value O* being an ideal output value of the AF sensor 28, calculates a value which is obtained by lower-limit guarding the detection absorption / discharge amount OSR# at the guard absorption / discharge amount OSRg during the execution of lean control, and a value which is obtained by upper-limit guarding the detection absorption / discharge amount OSR# at the guard absorption / discharge amount OSRg during the execution of rich control as corrected absorption / discharge amounts OSRc, and switches control to the lean control from the rich control, or to the rich control from the lean control when the corrected absorption / discharge amounts OSRc reach predefined switching absorption / discharge amounts OSR_LN, OSR_RC.SELECTED DRAWING: Figure 5A

Description

[Technical field]

[0001] This specification discloses a control device for an internal combustion engine that switches between rich control and lean control depending on the amount of oxygen adsorbed and released by a catalyst. [Background technology]

[0002] In general, a catalyst for purifying exhaust gas is provided in an exhaust passage of an internal combustion engine, and harmful substances (e.g., NOx, HC, CO) contained in the exhaust gas are purified by this catalyst. When the oxygen storage capacity of the catalyst is high, the purification efficiency of HC and CO increases, while the purification efficiency of NOx decreases. When the oxygen storage capacity of the catalyst is low, the purification efficiency of NOx increases, while the purification efficiency of HC and CO decreases. Therefore, a technique for alternately switching between lean control and rich control has been proposed in the past to appropriately adjust the oxygen storage capacity of the catalyst. In the lean control, the target air-fuel ratio of the internal combustion engine is set to a value (e.g., 14.65) higher than the theoretical air-fuel ratio (e.g., 14.6). In the rich control, the target air-fuel ratio is set to a value (e.g., 14.55) lower than the theoretical air-fuel ratio.

[0003] Patent Document 1 discloses a technique for estimating the increase or decrease in the amount of oxygen stored in a catalyst, and switching from lean control to rich control if the estimated increase or decrease reaches an upper threshold while lean control is being performed, and switching from rich control to lean control if the estimated increase or decrease reaches a lower threshold while rich control is being performed. In Patent Document 1, the estimated increase or decrease is calculated based on the detection value of a first air-fuel ratio sensor provided upstream of the catalyst, the fuel injection amount, and the oxygen concentration of the air. This technique can appropriately purify NOx, HC, and CO. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-090947 A Summary of the Invention [Problem to be solved by the invention]

[0005] As described above, the timing of switching between the lean control and the rich control is determined based on the change in the increase or decrease in the amount of oxygen storage. The increase or decrease is usually estimated from the detection value of the air-fuel ratio sensor. In this case, if the responsiveness of the air-fuel ratio sensor deteriorates, the accuracy of the estimated increase or decrease deteriorates accordingly. In this case, the switching period between the rich control and the lean control becomes longer, which may result in a deterioration in emissions.

[0006] Therefore, this specification discloses a control device for an internal combustion engine that can more effectively prevent the deterioration of emissions. [Means for solving the problem]

[0007] The control device for an internal combustion engine disclosed in this specification is a control device for an internal combustion engine that switches between rich control and lean control depending on the oxygen adsorption / release amount of a catalyst, and is characterized in that it is configured to calculate the oxygen adsorption / release amount as a detected adsorption / release amount based on an AF detection value, which is an actual output value of an air-fuel ratio sensor, calculate the oxygen adsorption / release amount as a guard adsorption / release amount based on an AF ideal value, which is an ideal output value of the air-fuel ratio sensor, calculate a value in which the detected adsorption / release amount is lower-limit guarded by the guard adsorption / release amount during an execution period of the lean control, and calculate a value in which the detected adsorption / release amount is upper-limit guarded by the guard adsorption / release amount during an execution period of the rich control, as a corrected adsorption / release amount, and switch from the rich control to the lean control, or from the lean control to the rich control, when the corrected adsorption / release amount reaches a predetermined switching adsorption / release amount.

[0008] Since the detected amount of oxygen adsorption / release is guarded at the upper or lower limit by the guard amount calculated from the ideal AF value, the corrected amount of oxygen adsorption / release after the guard process can be prevented from deviating significantly from the actual amount of oxygen adsorption / release. As a result, the cycle of switching between rich / lean control can be prevented from becoming excessively long, and the deterioration of emissions can be effectively prevented.

[0009] In this case, the oxygen adsorption / release amount may be calculated as an ideal adsorption / release amount based on the AF ideal value, and a predetermined post-correction process may be performed on the ideal adsorption / release amount to calculate the guard adsorption / release amount.

[0010] The post-correction process may include adding a specified correction value to the ideal absorption / release amount.

[0011] Furthermore, an ideal absorption / discharge amount may be calculated based on a value obtained by performing a predetermined pre-correction process on the AF ideal value, and the guard absorption / discharge amount may be calculated based on the ideal absorption / discharge amount.

[0012] In this case, the pre-correction process may include a smoothing process for converting the AF ideal value into a smooth waveform.

[0013] The pre-correction process may include changing a phase of the AF ideal value so that rising and falling points of the AF ideal value coincide with rising and falling points of the AF detection value.

[0014] By performing the post-correction process or the pre-correction process, the guard adsorption / release amount can be made closer to the actual oxygen adsorption / release amount, and deterioration of emissions can be prevented more effectively. Effect of the Invention

[0015] According to the technology disclosed in this specification, the deterioration of emissions can be prevented more effectively. [Brief description of the drawings]

[0016] [Figure 1] FIG. 2 is a block diagram showing the configuration of an engine. [Diagram 2] FIG. 4 is a diagram showing changes in the state of the catalyst when lean control and rich control are alternately switched. [Diagram 3] 11A and 11B are diagrams showing changes in the state of the catalyst when the response delay of the AF detection value becomes significantly large. [Figure 4]FIG. 4 is a diagram showing changes in the state of the catalyst when rich / lean control switching is performed according to the ideal absorption / release amounts. [Figure 5A] 11 is a block diagram showing a configuration of an LN switching determination unit. FIG. [Figure 5B] FIG. 4 is a block diagram showing a configuration of an RC switching determination unit. [Figure 6] FIG. 13 is a diagram for explaining the effect of guarding the detected amount of absorption / release with a lower limit or an upper limit. [Figure 7] 13 is a diagram illustrating an example of an LN switching determination unit when pre-correction is performed on an AF ideal value. FIG. [Figure 8] FIG. 13 is a diagram showing the effect of smoothing processing. [Figure 9] FIG. 13 is a diagram illustrating the effect of a phase shift process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, the control device 34 for an internal combustion engine will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an engine 10 that is the object of control by the control device 34. The engine 10 is, for example, an internal combustion engine that is mounted on a vehicle and outputs power. The engine 10 has a combustion chamber 12, an intake passage 18 connected to the combustion chamber 12, and an exhaust passage 24 connected to the combustion chamber 12. A throttle valve 20 is provided in the intake passage 18. Air is supplied to the combustion chamber 12 via the throttle valve 20 and the intake passage 18. A fuel injection valve 22 is also provided in the intake passage 18, and fuel is supplied to the combustion chamber 12 via the fuel injection valve 22 and the intake passage 18.

[0018] An ignition plug 16 is provided in the combustion chamber 12. When ignition is performed by the ignition plug 16, the air-fuel mixture supplied to the combustion chamber 12 is burned. Then, as the air-fuel mixture is burned in the combustion chamber 12, the piston 14 moves back and forth.

[0019] The exhaust gas generated after the mixture is burned is sent to the exhaust passage 24. A catalytic converter 26 is provided in the exhaust passage 24. The catalytic converter 26 has a three-way catalyst (hereinafter referred to as the "catalyst") 27 and purifies harmful components such as HC, CO, and NOx in the exhaust gas. A front air-fuel ratio sensor (hereinafter referred to as the "front AF sensor") 28 is provided upstream of the catalytic converter 26. The front AF sensor 28 outputs a linear signal according to the oxygen concentration in the exhaust gas upstream of the catalyst 27. Specifically, the AF detection value O#, which is the output value of the front AF sensor 28, becomes smaller as the oxygen concentration in the exhaust gas before passing through the catalyst 27 becomes thinner. In addition, the front AF sensor 28 is adjusted so that the AF detection value O# becomes a predetermined value Ost when the mixture is burned at the theoretical air-fuel ratio. In the following, this predetermined value Ost is called the "AF stoichiometric value Ost".

[0020] The control device 34 controls the operation of the engine 10. The control device 34 is physically a computer having a processor 36 and a memory 38. This "computer" also includes a microcontroller in which a computer system is incorporated into a single integrated circuit. The control device 34 may be composed of a single computer, or may be composed of multiple computers that are physically separated from one another.

[0021] The control device 34 controls the operation of the engine 10 according to the depression amount of the brake pedal 30 and the accelerator pedal 32. The control device 34 also estimates the oxygen adsorption / release amount OSR of the catalyst 27, and switches from lean control to rich control or from rich control to lean control according to the oxygen adsorption / release amount OSR. In the lean control, the control device 34 sets the target air-fuel ratio of the engine 10 to a value (e.g., 14.65) higher than the stoichiometric air-fuel ratio (e.g., 14.6). In the rich control, the control device 34 sets the target air-fuel ratio to a value (e.g., 14.55) lower than the stoichiometric air-fuel ratio. In either case of the lean control or the rich control, the control device 34 controls the opening degree of the throttle valve 20 and the fuel injection valve 22 so as to obtain a mixture having the set target air-fuel ratio.

[0022] Next, the state change of the catalyst 27 when lean control and rich control are alternately switched will be described with reference to Fig. 2. The first row of Fig. 2 shows the control state CS. When the control state CS is "LN", lean control is performed, and when the control state CS is "RC", rich control is performed.

[0023] The second row of FIG. 2 shows the AF detection value O#, which is the output value of the front AF sensor 28. As described above, the AF detection value O# is a value that corresponds to the oxygen concentration of the exhaust gas before it passes through the catalyst 27. Therefore, the AF detection value O# is an AF lean value Oln that is higher than the AF stoichiometric value Ost when lean control is being executed, and is an AF rich value Orc that is lower than the AF stoichiometric value Ost when rich control is being executed. However, the AF detection value O# contains a certain degree of response delay and noise. Therefore, the AF detection value O# is a dull signal compared to the control state CS.

[0024] The third row of FIG. 2 shows the detected absorption / release amount OSR#, which is the amount of oxygen absorption / release calculated based on the AF detection value O#. The detected absorption / release amount OSR# indicates the amount of oxygen released from the catalyst 27 or absorbed into the catalyst 27. This detected absorption / release amount OSR# is an integrated value of the instantaneous detected absorption / release amount ΔOSR#. The instantaneous detected absorption / release amount ΔOSR# is the amount of oxygen absorption / release per unit time, and is calculated by Equation 1 or Equation 2. In Equation 1 and Equation 2, Qa indicates the amount of fuel injection, and Aa indicates the amount of intake air. The detected absorption / release amount OSR# is reset to zero at the timing of switching between lean control and rich control. ΔOSR#=(O#-Ost)×Qa×0.23 Equation 1 ΔOSR#=(O#-Ost) / O#×Aa×0.23 Equation 2

[0025] As shown in the third row of FIG. 2, the detected absorption / discharge amount OSR# gradually increases during the execution period of lean control, and gradually decreases during the execution period of rich control. However, as described above, the AF detection value O# includes a certain degree of response delay. Therefore, immediately after the control is switched, the detected absorption / discharge amount OSR# increases or decreases in the opposite direction by the amount of this response delay. For example, immediately after switching from lean control to rich control (e.g., immediately after time t1), the detected absorption / discharge amount OSR# temporarily increases and then gradually decreases.

[0026] The control device 34 switches from lean control to rich control when the detected absorption and release amount OSR# becomes equal to or greater than the specified LN switchover absorption and release amount OSR_LN during execution of lean control. Also, the control device 34 switches from rich control to lean control when the detected absorption and release amount OSR# becomes equal to or less than the specified RC switchover absorption and release amount OSR_RC during rich control. The fourth row of FIG. 2 shows the oxygen storage amount OSA of the catalyst 27. As shown in FIG. 2, the oxygen storage amount OSA increases during lean control and decreases during rich control.

[0027] As described above, the reason for alternately switching between rich control and lean control is to efficiently purify the three harmful substances (i.e., HC, CO, and NOx) contained in the exhaust gas. That is, when the oxygen storage amount OSA of the catalyst 27 is low, the purification efficiency of NOx increases while the purification efficiency of HC and CO decreases. On the other hand, when the oxygen storage amount OSA of the catalyst 27 is high, the purification efficiency of HC and CO increases while the purification efficiency of NOx decreases. Therefore, in this example, the rich control and lean control are alternately switched so that the oxygen storage amount OSA repeatedly increases and decreases within a predetermined range.

[0028] As mentioned repeatedly, the AF detection value O# includes a response delay. When the front AF sensor 28 is normal, this response delay falls within a certain range. Therefore, the control device 34 sets the values ​​of various parameters (for example, the LN switching absorption / release amount OSR_LN and the RC switching absorption / release amount OSR_RC, etc.) on the premise that there is a response delay within a certain range. However, when the front AF sensor 28 deteriorates or a signal delay occurs for some reason, the response delay included in the AF detection value O# becomes excessively large. In this case, there is a risk that the rich / lean control switching period Tc becomes excessively large. This will be described with reference to FIG. 3.

[0029] Fig. 3 shows the state change of the catalyst 27 when the response delay of the AF detection value O# becomes significantly large. The first row of Fig. 3 shows the control state CS, the second row shows the AF detection value O#, and the third row shows the detected absorption / release amount OSR# calculated from the AF detection value O#. As shown in the second row of Fig. 3, in the example of Fig. 3, the response delay of the AF detection value O# becomes larger than in the example of Fig. 2, and the AF detection value O# reacts with a large delay from the timing of switching the rich / lean control.

[0030] Therefore, as shown in the third row of Fig. 3, after switching between rich and lean control, the detected absorption and release amount OSR# largely fluctuates in the opposite direction. As a result, the time until the detected absorption and release amount OSR# reaches the switching absorption and release amount OSR_RC, OSR_LN becomes longer. This lengthens the switching period Tc between the rich control and the lean control, which may cause the emission to deteriorate.

[0031] In order to avoid such a problem, the control device 34 of this embodiment calculates the oxygen adsorption / release amount as a guard adsorption / release amount OSRg based on the AF ideal value O*, which is the ideal output value of the front AF sensor 28, and corrects the detected adsorption / release amount OSR# using this guard adsorption / release amount OSRg. This will be described in detail below.

[0032] FIG. 4 shows the state change of the catalyst 27 when rich / lean control switching is performed according to the ideal intake / release amount OSR*. Specifically, the first row of FIG. 4 shows the control state CS, and the second row of FIG. 4 shows the AF ideal value O*. Here, the AF ideal value O* is the ideal output value of the front AF sensor 28. In this example, during the execution period of the rich control, O*=Orc is considered, and during the execution period of the lean control, O*=Oln is considered. As shown in FIG. 4, this AF ideal value O* follows the change in the control state CS without delay. Therefore, when the rich / lean control is switched, the value of the AF ideal value O* is also immediately switched between the AF rich value Orc and the AF rich value Orc.

[0033] The third row of FIG. 4 shows the ideal absorption / release amount OSR* calculated based on the AF ideal value O*. The ideal absorption / release amount OSR* is a value obtained by integrating the instantaneous ideal absorption / release amount ΔOSR*. The instantaneous ideal absorption / release amount ΔOSR* is calculated by replacing "O#" in formula 1 or formula 2 with "O*". In other words, the instantaneous ideal absorption / release amount ΔOSR* is calculated by the following formula 3 or formula 4. ΔOSR*=(O*-Ost)×Qa×0.23 Equation 3 ΔOSR*=(O*-Ost) / O*×Aa×0.23 Equation 4

[0034] As is clear from the third row of Fig. 4, the ideal intake / exhaust amount OSR* immediately rises or falls after the rich / lean control is switched. Therefore, the ideal intake / exhaust amount OSR* reaches the switching intake / exhaust amounts OSR_RC, OSR_LN in a relatively short time. In the example of Fig. 4, the rich / lean control is switched at the timing when the ideal intake / exhaust amount OSR* reaches the switching intake / exhaust amounts OSR_RC, OSR_LN.

[0035] The control device 34 of this embodiment uses this ideal absorption and release amount OSR* to prevent the rich / lean control switching period Tc from becoming excessively large. Figures 5A and 5B are block diagrams showing the configuration of the control switching determination unit 40, 60 using such ideal absorption and release amount OSR*. Figure 5A shows the LN switching determination unit 40 used during execution of lean control, and Figure 5B shows the RC switching determination unit 60 used during execution of rich control. In Figures 5A and 5B, these switching determination units 40, 60 are illustrated as circuits, but in reality, the functions of the switching determination units 40, 60 are realized in software by the control device 34 operating according to a predetermined program.

[0036] The LN switching determination unit 40 has a first calculator 42. The AF detection value O#, the fuel injection amount Qa, and the intake air amount Aa are input to the first calculator 42. The first calculator 42 calculates the instantaneous detected intake / release amount ΔOSR# according to Equation 1 or Equation 2. The instantaneous detected intake / release amount ΔOSR# is added to the detected intake / release amount OSR#[i-1] from one cycle before by a first adder 44. The output value from this first adder 44 becomes the detected intake / release amount OSR#.

[0037] Furthermore, the LN switching determination unit 40 has a second calculator 46. The AF ideal value O*, the fuel injection amount Qa, and the intake air amount Aa are input to the second calculator 46. The second calculator 46 calculates the instantaneous ideal intake and release amount ΔOSR* according to Equation 3 or Equation 4. The instantaneous ideal intake and release amount ΔOSR* is added to the ideal intake and release amount OSR*[i-1] of the previous cycle by a second adder 48. The output value from this second adder 48 becomes the ideal intake and release amount OSR*.

[0038] The ideal absorption / release amount OSR* is input to the post-correction unit 49. The post-correction unit 49 has a third adder 50. The third adder 50 adds the ideal absorption / release amount OSR* and a predetermined correction value Ka, and outputs the result as the guard absorption / release amount OSRg. The correction value Ka used in the LN switching determination unit 40 is a value that reduces the ideal absorption / release amount OSR*, that is, a negative value. This correction value Ka may be a predetermined fixed value, or may be a variable that varies depending on the control situation.

[0039] The lower limit guard processing unit 52 outputs the value obtained by lower limit guarding the detected absorption amount OSR# with the guard absorption amount OSRg as the corrected absorption amount OSRc. That is, when the detected absorption amount OSR# is less than the guard absorption amount OSRg, the lower limit guard processing unit 52 outputs the guard absorption amount OSRg as the corrected absorption amount OSRc. Also, when the detected absorption amount OSR# is greater than or equal to the guard absorption amount OSRg, the lower limit guard processing unit 52 outputs the detected absorption amount OSR# as the corrected absorption amount OSRc. In other words, the lower limit guard processing unit 52 outputs the higher value of the detected absorption amount OSR# and the guard absorption amount OSRg as the corrected absorption amount OSRc.

[0040] The comparator 54 compares the corrected absorption amount OSRc with a previously defined switching absorption amount OSR_LN for LN. Then, the comparator 54 outputs a switching signal Sg according to the comparison result. The switching signal Sg becomes High when OSRc ≥ OSR_LN, and becomes Low when OSRc < OSR_LN. When a High switching signal Sg is output, the control device 34 switches from lean control to rich control.

[0041] During the execution of rich control, as shown in FIG. 5B, the necessity of control switching is determined by the RC switching determination unit 60. The procedure for calculating the detected absorption amount OSR# and the guard absorption amount OSRg in the RC switching determination unit 60 is substantially the same as the procedure in the LN switching determination unit 40. However, the correction value Ka used in the RC switching determination unit 60 is a value that raises the ideal absorption amount OSR*, that is, a positive value.

[0042] The detected oxygen uptake amount OSR# and the guard oxygen uptake amount OSRg are input to the upper limit guard processing unit 72. The upper limit guard processing unit 72 outputs the value obtained by upper limit guarding the detected oxygen uptake amount OSR# with the guard oxygen uptake amount OSRg as the corrected oxygen uptake amount OSRc. That is, when the detected oxygen uptake amount OSR# is greater than or equal to the guard oxygen uptake amount OSRg, the upper limit guard processing unit 72 outputs the guard oxygen uptake amount OSRg as the corrected oxygen uptake amount OSRc. Also, when the detected oxygen uptake amount OSR# is less than the guard oxygen uptake amount OSRg, the upper limit guard processing unit 72 outputs the detected oxygen uptake amount OSR# as the corrected oxygen uptake amount OSRc. In other words, the upper limit guard processing unit 72 outputs the lower value of the detected oxygen uptake amount OSR# and the guard oxygen uptake amount OSRg as the corrected oxygen uptake amount OSRc.

[0043] The comparator 74 compares the corrected oxygen uptake amount OSRc with a pre-defined switching oxygen uptake amount OSR_RC for RC. Then, the comparator 74 outputs a switching signal Sg according to the comparison result. The switching signal Sg becomes High when OSRc < OSR_RC, and becomes Low when OSRc ≥ OSR_RC. If the High switching signal Sg is output, the control device 34 switches from rich control to lean control.

[0044] Next, with reference to FIG. 6, the effects when the detected oxygen uptake amount OSR# is lower limit or upper limit guarded with the guard oxygen uptake amount OSRg will be described. FIG. 6 is a diagram showing an example of the oxygen uptake amount OSR obtained during lean control. In FIG. 6, the broken line indicates the AF detection value O#_n and the detected oxygen uptake amount OSR#_n obtained when the front AF sensor 28 is normal. Also, the dashed-dotted line indicates the AF detection value O#_a and the detected oxygen uptake amount OSR#_a obtained when the response delay of the front AF sensor 28 excessively increases. Further, the thick solid line indicates the AF ideal value O*, and the guard oxygen uptake amount OSRg calculated based on the AF ideal value O*.

[0045] The guard absorption / discharge amount OSRg is a value obtained by adding a correction value Ka to the ideal absorption / discharge amount OSR*. In the example of Fig. 6, the correction value Ka is set to a value that makes the guard absorption / discharge amount OSRg smaller than the normal detected absorption / discharge amount OSR#_n. Therefore, when the front AF sensor 28 is normal, the detected absorption / discharge amount OSR# is output as the corrected absorption / discharge amount OSRc. Then, at the point in time when the detected absorption / discharge amount OSR#_n (i.e., the corrected absorption / discharge amount OSRc) reaches the LN switching absorption / discharge amount OSR_LN (i.e., time t2 in Fig. 6), the control switches to rich control.

[0046] On the other hand, when the response delay of the front AF sensor 28 becomes large, the guard absorption / discharge amount OSRg becomes larger than the detected absorption / discharge amount OSR#_a. In this case, the guard absorption / discharge amount OSRg is output as the corrected absorption / discharge amount OSRc. Then, even if the detected absorption / discharge amount OSR#_a has not reached the LN switching absorption / discharge amount OSR_LN, the control switches to rich control at the time when the guard absorption / discharge amount OSRg (i.e., the corrected absorption / discharge amount OSRc) reaches the LN switching absorption / discharge amount OSR_LN (i.e., time t3 in FIG. 6). This prevents the rich / lean control switching period Tc from becoming excessively large. As a result, the deterioration of emissions can be effectively prevented.

[0047] In this example, the ideal absorption / discharge amount OSR* is subjected to post-correction processing in which the correction value Ka is added to the ideal absorption / discharge amount OSR*, and the result is calculated as the guard absorption / discharge amount OSRg. However, the guard absorption / discharge amount OSRg may be another value as long as it depends on the AF ideal value O*. For example, the guard absorption / discharge amount OSRg may be the ideal absorption / discharge amount OSR* itself calculated by applying the AF ideal value O* to formula 3 or formula 4.

[0048] In addition, a predetermined pre-correction process may be applied to the AF ideal value O* prior to the calculation of the ideal absorption and release amount OSR*. FIG. 7 is a diagram showing an example of the LN switching determination unit 40 when the AF ideal value O* is subjected to a pre-correction process. As shown in FIG. 7, the LN switching determination unit 40 has a pre-correction unit 41. The pre-correction unit 41 performs a predetermined pre-correction process on the AF ideal value O*. The first calculator 42 calculates the ideal absorption and release amount OSR* based on the corrected AF ideal value O**. The subsequent processes are the same as those in FIG. 5A, and therefore the description will be omitted. Although FIG. 7 shows a post-correction unit 49, when the pre-correction unit 41 is provided, the post-correction unit 49 may be omitted.

[0049] Here, the pre-correction process performed by the pre-correction unit 41 is not particularly limited as long as it is a process that brings the ideal absorption / discharge amount OSR* closer to the detected absorption / discharge amount OSR#_n in a normal state. Therefore, for example, the pre-correction process may be a smoothing process that converts the AF ideal value O* into a smooth waveform. For example, the pre-correction unit 41 may output a moving average value of the AF ideal value O* as the corrected AF ideal value O**.

[0050] Fig. 8 is a diagram showing the effect of the smoothing process. In Fig. 8, the thick solid line indicates the AF ideal value O* before correction and the ideal absorption and release amount OSR* calculated from this AF ideal value O*. In Fig. 8, the dashed line indicates the AF ideal value O** after smoothing process and the ideal absorption and release amount OSR** calculated from this AF ideal value O** after correction.

[0051] As is clear from Fig. 8, by performing the smoothing process, the behavior of the ideal absorption / release amount OSR** becomes closer to the detected absorption / release amount OSR# in normal times (i.e., the detected absorption / release amount OSR# in Fig. 2). By using such ideal absorption / release amount OSR** as the guard absorption / release amount OSRg, it is possible to prevent the rich / lean control switching period Tc from deviating significantly from the normal switching period Tc.

[0052] As another embodiment, the pre-correction process may be a process of shifting the phase of the AF ideal value O* as shown in Fig. 9. For example, the pre-correction unit 41 may shift the phase of the AF ideal value O* so that the rising point Pup and the falling point Pdn of the AF ideal value O** coincide with the rising point Pup and the falling point Pdn of the AF detection value O#.

[0053] Fig. 9 is a diagram showing the effect of the phase shift processing. In Fig. 9, the first row shows the control state CS, and the second row shows the AF detection value O#. The thick solid lines in the third and fourth rows of Fig. 9 show the AF ideal value O* before correction and the ideal absorption and release amount OSR* calculated from this AF ideal value O*. The dashed lines in the third and fourth rows of Fig. 9 show the AF ideal value O** after the phase shift processing and the ideal absorption and release amount OSR** calculated from this AF ideal value O** after correction.

[0054] As is clear from Fig. 9, by performing the phase shift processing, the behavior of the ideal intake / discharge amount OSR** becomes closer to the detected intake / discharge amount OSR# in normal times (i.e., the detected intake / discharge amount OSR# in Fig. 2). By using such ideal intake / discharge amount OSR** as the guard intake / discharge amount OSRg, it is possible to prevent the rich / lean control cycle from deviating significantly from the normal control cycle.

[0055] In addition, the above-described addition process of the correction value Ka, the smoothing process of the AF ideal value O*, and the phase shift process of the AF ideal value O* may be combined with each other. In another embodiment, neither the pre-correction process nor the post-correction process may be performed. Furthermore, the contents of the pre-correction process and the post-correction process may be changed as appropriate. For example, the post-correction process may be a smoothing process that smoothes the waveform of the calculated ideal absorption / release amount OSR*. In addition, the pre-correction process may be an addition process that adds a predetermined correction value to the AF ideal value O*.

[0056] In this example, the predefined AF lean value Oln and AF rich value Orc are used as the AF ideal value O*. However, the ratio between the fuel injection amount Qa and the intake air amount Aa may be used as the AF ideal value O*. That is, O*=Aa / Qa may be used. [Explanation of symbols]

[0057] 10 engine, 12 combustion chamber, 14 piston, 16 spark plug, 18 intake passage, 20 throttle valve, 22 fuel injector, 24 exhaust passage, 26 catalytic converter, 27 catalyst, 28 front AF sensor, 30 brake pedal, 32 accelerator pedal, 34 control device, 36 processor, 38 memory, 40 LN switching determination unit, 41 pre-correction unit, 42 first calculator, 44 first adder, 46 second calculator, 48 second adder, 49 post-correction unit, 50 third adder, 52 lower limit guard processing unit, 54 comparator, 60 RC switching determination unit, 62 first calculator, 72 upper limit guard processing unit, 74 comparator.

Claims

1. A control device for an internal combustion engine that switches between rich control and lean control depending on an amount of oxygen adsorbed and released by a catalyst, Calculating the amount of oxygen adsorption / release as a detected amount of oxygen adsorption / release based on an AF detection value, which is an actual output value of an air-fuel ratio sensor; An oxygen adsorption / release amount is calculated as an ideal adsorption / release amount based on an AF ideal value which is an ideal output value of the air-fuel ratio sensor, and a value obtained by performing a predetermined post-correction process on the ideal adsorption / release amount is calculated as a guard adsorption / release amount; A value obtained by guarding the lower limit of the detected amount of adsorption / release with the guard amount of adsorption / release during the execution period of the lean control, and a value obtained by guarding the upper limit of the detected amount of adsorption / release with the guard amount of adsorption / release during the execution period of the rich control are calculated as the corrected amount of adsorption / release, When the corrected absorption / release amount reaches a predetermined switching absorption / release amount, the rich control is switched to the lean control, or the lean control is switched to the rich control, The post-correction process is a process of adding a specified correction value to the ideal absorption / release amount, The correction value is a negative value during the execution of the lean control, and is a positive value during the execution of the rich control, The correction value is a value calculated as the corrected absorption / release amount when the air-fuel ratio sensor is normal. A control device for an internal combustion engine, comprising:

2. 2. The control device for an internal combustion engine according to claim 1, 2. A control device for an internal combustion engine, comprising: a controller for controlling an internal combustion engine, the controller being operable to calculate the ideal intake and release amount based on a value obtained by performing a predetermined pre-correction process on the AF ideal value.

3. The control device for an internal combustion engine according to claim 2, 4. The control device for an internal combustion engine, wherein the pre-correction process includes a smoothing process for converting the AF ideal value into a smooth waveform.

4. The control device for an internal combustion engine according to claim 2, The control device for an internal combustion engine, characterized in that the pre-correction process includes changing a phase of the AF ideal value so that rising points and falling points of the AF ideal value coincide with rising points and falling points of the AF detection value.

Images