Control device for internal combustion engines

Abstract

To enable a three-way catalyst to exert high exhaust emission control performance regardless of the degree of deterioration of the three-way catalyst.SOLUTION: A control device 80 can execute: processing of calculating a feedback value for matching the air-fuel ratio and the theoretical air-fuel ratio of exhaust gas on a downstream side of a first catalyst 32; processing of calculating a catalyst correction value such that a target air-fuel ratio in a cylinder 12 is corrected to the richer side as an intake air amount increases; processing of calculating the target air-fuel ratio based on the feedback value and the catalyst correction value; processing of controlling a fuel injection amount of an injector 22 so that the air-fuel ratio of exhaust gas on an upstream side of the first catalyst 32 matches with the target air-fuel ratio; and processing of calculating the degree of deterioration of the first catalyst 32. Assuming that the intake air amount is the same, in the processing of calculating the catalyst correction value, when the degree of deterioration of the first catalyst 32 is a first deterioration degree, the catalyst correction value is calculated so that the amount of correction to the richer side is greater than that when the first catalyst 32 is at a second deterioration degree that is lower than the first deterioration degree.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] This invention relates to a control device for an internal combustion engine.

Background Art

[0002] The internal combustion engine disclosed in Patent Document 1 includes a three-way catalyst disposed in an exhaust passage, an upstream air-fuel ratio sensor located upstream of the three-way catalyst in the exhaust passage, a downstream air-fuel ratio sensor located downstream of the three-way catalyst in the exhaust passage, and an injector. The control device for this internal combustion engine controls the fuel injection amount of the injector. At this time, the control device adjusts the air-fuel ratio of the exhaust gas flowing into the three-way catalyst to be near the theoretical air-fuel ratio through feedback control based on the detection value of the upstream air-fuel ratio sensor and feedback control based on the detection value of the downstream air-fuel ratio sensor.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a technology such as Patent Document 1 that controls the air-fuel ratio so as to obtain high exhaust gas purification performance with a three-way catalyst, the three-way catalyst may deteriorate. When the three-way catalyst deteriorates, even if the air-fuel ratio of the exhaust gas flowing into the three-way catalyst is close to the theoretical air-fuel ratio, the three-way catalyst may not be able to exhibit high exhaust gas purification performance. In Patent Document 1, there is no examination on how to make the three-way catalyst exhibit high exhaust gas purification performance when it deteriorates. Therefore, there is room for examination in this regard.

Means for Solving the Problems

[0005] The control device for an internal combustion engine to solve the above problems is applied to an internal combustion engine comprising: a three-way catalytic converter located in the exhaust passage; an upstream sensor that detects the upstream air-fuel ratio, which is the air-fuel ratio of the exhaust upstream of the three-way catalytic converter in the exhaust passage; a downstream sensor that detects the downstream air-fuel ratio, which is the air-fuel ratio of the exhaust downstream of the three-way catalytic converter in the exhaust passage; an air flow meter that detects the intake air volume; and an injector that injects fuel to be supplied into the cylinder. Based on the difference between the downstream air-fuel ratio and the stoichiometric air-fuel ratio, the control device detects the downstream air-fuel ratio of Theoretical air-fuel ratio to The following processes can be executed: a process to calculate a feedback value for matching; a process to calculate a catalyst correction value so that the target air-fuel ratio in the cylinder is corrected to the rich side as the intake air volume increases; a process to calculate the target air-fuel ratio based on the feedback value and the catalyst correction value; a process to control the fuel injection amount of the injector so that the upstream air-fuel ratio matches the target air-fuel ratio; and a process to calculate the degree of deterioration of the three-way catalyst. Assuming the intake air volume is the same, in the process to calculate the catalyst correction value, the degree of deterioration of the three-way catalyst is The higher The catalyst correction value is calculated such that the correction amount on the rich side is increased. [Effects of the Invention]

[0006] With the above technical concept, it is possible to enable the three-way catalyst to exhibit high exhaust gas purification performance regardless of the degree of deterioration of the three-way catalyst. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a schematic diagram of an internal combustion engine. [Figure 2] Figure 2 is a block diagram showing each process related to injector control. [Figure 3] Figure 3 is a graph representing the first map. [Figure 4] Figure 4 is a graph representing the second map. [Figure 5] Figure 5 schematically illustrates the process of oxygen adsorption and desorption in the first catalyst. [Modes for carrying out the invention]

[0008] <Overall structure of an internal combustion engine> Hereinafter, one embodiment of a control device for an internal combustion engine will be described with reference to the drawings. As shown in Figure 1, the vehicle 100 is equipped with an internal combustion engine 10. The internal combustion engine 10 is equipped with a plurality of cylinders 12. Each cylinder 12 is a space partitioned within the engine body 10A for burning a mixture of fuel and intake air. A crankshaft (not shown) rotates in response to the combustion of the mixture in the cylinders 12.

[0009] The internal combustion engine 10 has multiple spark plugs 19. One spark plug 19 is provided for each cylinder 12. The spark plugs 19 ignite the fuel-air mixture in the cylinder 12. The internal combustion engine 10 comprises an intake passage 20, a throttle valve 21, and a plurality of injectors 22. The intake passage 20 is a passage for introducing intake air into each cylinder 12. The intake passage 20 is connected to each cylinder 12. The throttle valve 21 is located in the middle of the intake passage 20. The throttle valve 21 adjusts the amount of intake air GA. An injector 22 is provided for each cylinder 12. The injector 22 is located downstream of the throttle valve 21 in the intake passage 20. The injector 22 injects fuel. The fuel injected by the injector 22 reaches the cylinder 12 via the intake passage 20. That is, the injector 22 injects fuel to be supplied into the cylinder 12.

[0010] The internal combustion engine 10 includes an exhaust passage 30, a first catalyst 32, and a second catalyst 34. The exhaust passage 30 is a passage for discharging exhaust gas from each cylinder 12. The exhaust passage 30 is connected to each cylinder 12. The first catalyst 32 is located in the middle of the exhaust passage 30. The first catalyst 32 is a three-way catalyst. That is, the first catalyst 32 purifies the exhaust gas by oxidizing HC and CO in the exhaust gas and reducing NOx in the exhaust gas when the air-fuel ratio of the exhaust gas is near the stoichiometric air-fuel ratio. The first catalyst 32 also has an oxygen storage function, which involves absorbing oxygen in the exhaust gas when the air-fuel ratio of the exhaust gas passing through the first catalyst 32 is leaner than the stoichiometric air-fuel ratio, and releasing the absorbed oxygen when the air-fuel ratio is richer than the stoichiometric air-fuel ratio. The second catalyst 34 is located downstream of the first catalyst 32 in the exhaust passage 30. The second catalyst 34 is a three-way catalyst, similar to the first catalyst 32.

[0011] The internal combustion engine 10 is equipped with an air flow meter 61, an upstream sensor 62, and a downstream sensor 63. The air flow meter 61 is located upstream of the throttle valve 21 in the intake passage 20. The air flow meter 61 detects the intake air volume GA. The upstream sensor 62 is located upstream of the first catalyst 32 in the exhaust passage 30. The upstream sensor 62 detects the upstream air-fuel ratio AFf. The upstream air-fuel ratio AFf is the air-fuel ratio of the exhaust gas flowing upstream of the first catalyst 32 in the exhaust passage 30. The downstream sensor 63 is located between the first catalyst 32 and the second catalyst 34 in the exhaust passage 30. The downstream sensor 63 detects the downstream air-fuel ratio AFr. The downstream air-fuel ratio AFr is the air-fuel ratio of the exhaust gas flowing downstream of the first catalyst 32. The upstream sensor 62 and the downstream sensor 63 are well-known limiting current type sensors and exhibit a linear response to the exhaust air-fuel ratio.

[0012] <Control device> Vehicle 100 is equipped with a control device 80. The control device 80 comprises a CPU 82 and a memory 84. The memory 84 pre-stores a program that describes the processing to be executed by the CPU 82. The CPU 82 controls the internal combustion engine 10 by executing the program stored in the memory 84. The control device 80 repeatedly receives the intake air volume GA detected by the air flow meter 61, the upstream air-fuel ratio AFf detected by the upstream sensor 62, and the downstream air-fuel ratio AFr detected by the downstream sensor 63 from each sensor.

[0013] As shown in Figure 2, the CPU 82 controls the fuel injection amount of the injector 22 basically through normal control U while the ignition switch of the vehicle 100 is on. That is, the CPU 82 basically continues normal control U during the above period. Depending on the situation, the CPU 82 may also cancel normal control U and control the fuel injection amount of the injector 22 by other controls. In any of the controls, the CPU 82 sets a target air-fuel ratio AFt in the cylinder 12 and adjusts the fuel injection amount of the injector 22 to achieve the set target air-fuel ratio AFt. Below, we will first explain one of the other controls, specific control V. Then, we will explain normal control U.

[0014] <Specific control> Specific control V is a dedicated control for executing the storage amount calculation process V1. The storage amount calculation process V1 is a process for calculating the maximum storage amount Cmax in the first catalyst 32. The maximum storage amount Cmax is the maximum amount of oxygen that the first catalyst 32 can absorb. The maximum storage amount Cmax is an indicator of the degree of degradation of the first catalyst 32. The smaller the maximum storage amount Cmax, the higher the degree of degradation of the first catalyst 32.

[0015] After the ignition switch is turned on and the operating state of the internal combustion engine 10 stabilizes, the CPU 82 performs specific control V. In specific control V, the CPU 82 performs the following: First, the CPU 82 sets the target air-fuel ratio AFt in cylinder 12 to be richer than the stoichiometric air-fuel ratio and injects fuel into the injector 22. When this fuel burns in cylinder 12, rich exhaust reaches the first catalyst 32. The first catalyst 32 then releases oxygen. Eventually, when all the oxygen absorbed by the first catalyst 32 is released, the exhaust no longer contains oxygen, and the downstream air-fuel ratio AFr detected by the downstream sensor 63 becomes rich. When the downstream air-fuel ratio AFr becomes rich, the CPU 82 sets the target air-fuel ratio AFt in cylinder 12 to be leaner than the stoichiometric air-fuel ratio and injects fuel into the injector 22. As a result, lean exhaust is discharged from cylinder 12. Simultaneously, the first catalyst 32 absorbs oxygen contained in the exhaust gas. Eventually, when the oxygen absorption capacity of the first catalyst 32 reaches its limit, the oxygen contained in the exhaust gas passes through the first catalyst 32 without being absorbed. Consequently, the downstream air-fuel ratio AFr detected by the downstream sensor 63 becomes lean. During this process, the CPU 82 performs the oxygen absorption amount calculation process V1. Specifically, the CPU 82 calculates the maximum oxygen absorption amount Cmax by accumulating the amount of oxygen that has flowed into the first catalyst 32 from the moment the target air-fuel ratio AFt is reversed from rich to lean until the downstream air-fuel ratio AFr becomes lean. Once the CPU 82 has calculated the maximum oxygen absorption amount Cmax, it stores the calculated maximum oxygen absorption amount Cmax in the memory 84. These processes of calculating the maximum oxygen absorption amount Cmax and storing the said maximum oxygen absorption amount Cmax constitute the oxygen absorption amount calculation process V1. After completing the oxygen absorption amount calculation process V1, the CPU 82 terminates the specific control V. Furthermore, memory 84 will continue to store the maximum absorbed amount Cmax calculated in the absorbed amount calculation process V1 until the CPU 82 next executes the absorbed amount calculation process V1, that is, until the ignition switch is turned on and the CPU 82 executes the specific control V. The absorbed amount calculation process V1 is a process that calculates the degree of degradation of the first catalyst 32.

[0016] <Normal control> As shown in FIG. 2, during the execution of the normal control U, the CPU 82 repeatedly performs each of the sub-feedback process U1, the learning process U2, the characteristic process U3, the target calculation process U4, and the main feedback process U5. The content of each of these processes will be described in detail below. Although a step-by-step explanation is omitted, the CPU 82 appropriately stores the values of the parameters calculated in each process in the memory 84.

[0017] <Sub-feedback process> The CPU 82 repeatedly executes the sub-feedback process U1 at a predetermined control cycle. In the sub-feedback process U1, the CPU 82 calculates a feedback value FB. The feedback value FB is a parameter for making the downstream air-fuel ratio AFr coincide with the theoretical air-fuel ratio. The CPU 82 calculates this feedback value FB based on the difference between the downstream air-fuel ratio AFr and the theoretical air-fuel ratio. Specifically, the CPU 82 calculates the feedback value FB by applying PID control to the air-fuel ratio deviation ΔAF obtained by subtracting the theoretical air-fuel ratio from the downstream air-fuel ratio AFr detected by the downstream sensor 63. That is, the CPU 82 calculates the latest feedback value FB as the sum of the proportional term, the integral term, and the differential term based on the air-fuel ratio deviation ΔAF. The proportional term is a value obtained by multiplying the current air-fuel ratio deviation ΔAF by a predetermined first coefficient. The integral term is a value obtained by multiplying the integral value of the air-fuel ratio deviation ΔAF up to now by a predetermined second coefficient. The differential term is a value obtained by multiplying the differential value of the air-fuel ratio deviation ΔAF based on the difference between the current air-fuel ratio deviation ΔAF and the previous air-fuel ratio deviation ΔAF by a predetermined third coefficient. The CPU 82 calculates the feedback value FB once for each execution of the sub-feedback process U1.

[0018] Incidentally, for example, when the upstream air-fuel ratio AFf is richer than the stoichiometric air-fuel ratio, and as a result rich exhaust blows through downstream of the first catalyst 32, the downstream air-fuel ratio AFr becomes richer than the stoichiometric air-fuel ratio. And when the downstream air-fuel ratio AFr is rich, the above-described air-fuel ratio deviation ΔAF becomes a negative value. In this case, basically, the CPU 82 updates the feedback value FB to the negative side. Along with this, the target air-fuel ratio AFt calculated by (Equation 1) described later becomes larger than the previous value. That is, the target air-fuel ratio AFt is corrected to the lean side. This acts in the direction of reducing the fuel injection amount of the injector 22. As a result, the upstream air-fuel ratio AFf and thus the downstream air-fuel ratio AFr are corrected to the lean side. And the downstream air-fuel ratio AFr approaches the stoichiometric air-fuel ratio. Thus, the feedback value FB acts so as to match the downstream air-fuel ratio AFr with the stoichiometric air-fuel ratio.

[0019] <Learning Process> The CPU 82 executes the learning process U2 each time a predetermined update timing arrives. The update timing arrives, for example, when the CPU 82 executes the sub-feedback process U1 a plurality of times.

[0020] The learning process U2 includes pre-processing and main processing. When the learning process U2 is started, the CPU 82 first performs pre-processing. In pre-processing, the CPU 82 determines whether a predetermined learning condition is met. The learning condition is that the intake air volume GA is less than or equal to the judgment value GAx. In determining whether the learning condition is met, the CPU 82 first calculates the judgment value GAx. As a prerequisite for the CPU 82 to calculate the judgment value GAx, the memory 84 has already stored the first map. As shown in Figure 3, the first map represents the relationship between the judgment value GAx and the maximum storage amount Cmax. In the first map, the judgment value GAx is predetermined as a value according to the magnitude of the maximum storage amount Cmax. Specifically, in the first map, if the maximum storage amount Cmax is less than the reference value CS, the judgment value GAx is the first value G1. On the other hand, in the first map, if the maximum storage amount Cmax is greater than or equal to the reference value CS, the judgment value GAx is the second value G2. The first value G1 is smaller than the second value G2. The first value G1 is predetermined, for example, through experiments, as the maximum intake air volume GA that allows updating the learned value Y when the degradation level of the first catalyst 32 is reasonably high. The second value G2 is predetermined, for example, through experiments, as the maximum intake air volume GA that allows updating the learned value Y when the degradation level of the first catalyst 32 is reasonably low. The reference value CS is predetermined, for example, through experiments, as the optimal threshold for switching the judgment value GAx used in pretreatment between the first value G1 and the second value G2. The CPU 82 uses this first map to calculate the judgment value GAx. That is, in calculating the judgment value GAx, the CPU 82 refers to the first map and the latest maximum storage amount Cmax calculated in the storage amount calculation process V1. Then, the CPU 82 calculates the judgment value GAx corresponding to the maximum storage amount Cmax in the first map as the judgment value GAx to be used in this pretreatment. Reflecting the contents of the first map described above, CPU82 sets the judgment value GAx to a smaller value when the latest maximum absorption amount Cmax is less than the reference value CS, compared to when the maximum absorption amount Cmax is equal to or greater than the reference value CS. In other words, CPU82 sets the judgment value GAx to a smaller value when the degree of degradation of the first catalyst 32 is higher than the reference degree, compared to when the degree of degradation of the first catalyst 32 is less than or equal to the reference degree.The standard degree is the degree of deterioration corresponding to the standard value CS.

[0021] CPU82 calculates a judgment value GAx and compares it with the current intake air volume GA detected by the air flow meter 61. If the intake air volume GA is less than or equal to the judgment value GAx, CPU82 determines that the learning condition is met and proceeds with the main processing of the learning process U2. On the other hand, if the intake air volume GA is greater than the judgment value GAx, CPU82 determines that the learning condition is not met and cancels the main processing.

[0022] In this process, CPU82 updates the learned value Y according to the feedback value FB. Specifically, CPU82 updates the learned value Y by transferring the integral term component from the feedback value FB to the learned value Y. That is, CPU82 calculates a new learned value Y by adding the current integral term to the learned value Y calculated in the previous learning process U2, while resetting the integral term to zero. In this way, CPU82 updates the learned value Y once in each run of this process.

[0023] Here, the feedback value FB is a superposition of various fluctuating elements related to the execution of the main feedback process U5, which will be described later. The integral term in the feedback value FB reflects the constant error component inherent in the air-fuel ratio control among the various fluctuating elements. When this integral term is transferred from the feedback value FB to the learned value Y, the feedback value FB mainly contains only the temporary fluctuating component. At the same time, when the feedback value FB is viewed as a time series, the feedback value FB will fluctuate around zero. In other words, by absorbing the integral term into the learned value Y, the feedback value FB can be brought closer to zero. Thus, the learned value Y is a parameter for bringing the feedback value FB closer to zero. The constant error component reflected in the integral term is due to, for example, the aging deterioration and individual differences of engine parts related to air-fuel ratio control, such as the upstream sensor 62. In other words, the learned value Y can be said to be a value for compensating for deviations in the air-fuel ratio caused by irreversible changes in the internal combustion engine 10, such as the aging deterioration and individual differences of the internal combustion engine 10.

[0024] <Characterization> As shown in Figure 2, the CPU 82 repeatedly executes the characteristic processing U3 at a predetermined control cycle. In the characteristic processing U3, the CPU 82 calculates the catalyst correction value H. The catalyst correction value H is a parameter for correcting the target air-fuel ratio AFt in the cylinder 12 according to the intake air amount GA. In conjunction with (Equation 1) described later, the larger the value of the catalyst correction value H, the more it contributes to shifting the target air-fuel ratio AFt towards the rich side. In other words, the catalyst correction value H is a parameter for correcting the target air-fuel ratio AFt towards the rich side according to the intake air amount GA. As a prerequisite for the CPU 82 to calculate this catalyst correction value H, the memory 84 has already stored a second map. As shown in Figure 4, the second map represents the relationship between the intake air amount GA and the catalyst correction value H for various maximum storage amounts Cmax in a Cartesian coordinate system with the intake air amount GA as the X axis and the catalyst correction value H as the Y axis. Note that in Figure 4, the relationship between the intake air amount GA and the catalyst correction value H is shown only for three of the various maximum storage amounts Cmax. The three maximum absorption capacities Cmax in Figure 4 have the relationship "first absorption capacity C1 < second absorption capacity C2 < third absorption capacity C3". As shown in Figure 4, the intake air volume GA, the catalyst correction value H, and the maximum absorption capacity Cmax have the following relationship. When considering the same maximum absorption capacity Cmax, the larger the intake air volume GA, the larger the catalyst correction value H. Also, when considering the same intake air volume GA, the smaller the maximum absorption capacity Cmax, that is, the higher the degree of degradation of the first catalyst 32, the larger the catalyst correction value H. The CPU 82 calculates the catalyst correction value H based on this second map. Specifically, the CPU 82 calculates the catalyst correction value H corresponding to the current intake air volume GA detected by the air flow meter 61 and the latest maximum absorption capacity Cmax calculated in the absorption capacity calculation process V1, based on the second map. Reflecting the contents of the second map, the CPU 82 calculates the catalyst correction value H as follows. Assuming the maximum absorption capacity Cmax is the same, the CPU 82 calculates a larger catalyst correction value H as the intake air volume GA increases. Also, assuming the intake air volume GA is the same, the CPU 82 calculates a larger catalyst correction value H as the maximum absorption capacity Cmax decreases, that is, as the degree of degradation of the first catalyst 32 increases.As described above, the larger the catalyst correction value H, the greater the correction amount towards the rich side of the target air-fuel ratio AFt. In other words, the CPU 82 calculates the catalyst correction value H such that the larger the intake air volume GA, or the higher the degree of deterioration of the first catalyst 32, the greater the correction amount towards the rich side of the target air-fuel ratio AFt. In this way, the CPU 82 calculates the catalyst correction value H once for each characteristic processing U3.

[0025] Here, the range of air-fuel ratios in which the exhaust purification function of the first catalyst 32 is high is called the catalyst window. The catalyst window has the characteristic of shifting towards the rich side as the intake air volume GA increases. Since the catalyst window is dependent on the intake air volume GA, in order to maintain the exhaust purification performance of the first catalyst 32, it is necessary to change the target air-fuel ratio AFt in cylinder 12 in accordance with the change in intake air volume GA. Now, suppose the intake air volume GA changes abruptly. And suppose the catalyst window also changes abruptly as a result. When the intake air volume GA changes abruptly, the feedback value FB responds, and eventually the target air-fuel ratio AFt can be brought within the catalyst window. However, it takes a considerable amount of time for the feedback value FB to reach an appropriate value. The catalyst correction value H is a parameter that functions effectively in quickly changing the target air-fuel ratio AFt in response to changes in intake air volume GA.

[0026] <Target calculation process> As shown in Figure 2, the CPU 82 repeatedly executes the target calculation process U4 at a predetermined control cycle. In the target calculation process U4, the CPU 82 calculates the target air-fuel ratio AFt in cylinder 12 based on the latest feedback value FB calculated in the sub-feedback process U1, the latest learned value Y calculated in the learning process U2, and the latest catalyst correction value H calculated in the characteristic process U3. Specifically, the CPU 82 calculates the target air-fuel ratio AFt as the value obtained by dividing the reference air-fuel ratio AFb by the target adjustment value, as shown in (Equation 1) below. The reference air-fuel ratio AFb is, for example, the stoichiometric air-fuel ratio. The target adjustment value is the sum of "1", the feedback value FB, the learned value Y, and the catalyst correction value H. (Equation 1) AFt = AFb / (1 + FB + Y + H) CPU82 calculates the target air-fuel ratio AFt once per target air-fuel ratio calculation process.

[0027] <Main Feedback Processing> The CPU 82 repeatedly executes the main feedback process U5 at a predetermined control cycle. In each main feedback process U5, the CPU 82 performs the following processing: Specifically, in the main feedback process U5, the CPU 82 controls the fuel injection amount of the injector 22 so that the upstream air-fuel ratio AFf matches the target air-fuel ratio AFt. More precisely, the CPU 82 corrects the fuel injection amount relative to the previous injection amount, which was calculated in the previous main feedback process U5, according to the difference between the current upstream air-fuel ratio AFf detected by the upstream sensor 62 and the target air-fuel ratio AFt. If the upstream air-fuel ratio AFf is leaner than the target air-fuel ratio AFt, the CPU 82 increases the fuel injection amount relative to the previous injection amount. On the other hand, if the upstream air-fuel ratio AFf is richer than the target air-fuel ratio AFt, the CPU 82 decreases the fuel injection amount relative to the previous injection amount. Once the CPU 82 calculates the new fuel injection amount, it injects this amount of fuel into the injector 22.

[0028] <Operation of the Embodiment> During operation of the internal combustion engine 10, the first catalyst 32 takes in oxygen from the exhaust gas and releases oxygen into the exhaust gas. For the first catalyst 32 to exhibit high exhaust gas purification performance, the amount of oxygen absorbed and released must be in equilibrium. The catalyst window described above is the range of air-fuel ratios in which this equilibrium is achieved.

[0029] Here, as shown in Figure 5, the first catalyst 32 comprises a support 32a and a noble metal 32b disposed on the surface of the support 32a. The support 32a is made of, for example, cerium oxide. The noble metal 32b is, for example, platinum. There are multiple types of noble metals 32b. When the first catalyst 32 takes in oxygen from the exhaust gas, it takes in oxygen directly into the support 32a without going through the noble metal 32b. On the other hand, when the first catalyst 32 releases the oxygen it has absorbed, it requires a reaction with the noble metal 32b. In relation to these factors, oxygen intake occurs rapidly in the first catalyst 32, but oxygen release takes time.

[0030] In situations where oxygen release takes longer than oxygen intake, it is necessary to create conditions favorable to oxygen release in order to balance the amount of oxygen absorbed and desorbed by the first catalyst 32. This requirement becomes more pronounced when the intake air volume GA is large, that is, when the amount of oxygen absorbed and desorbed per unit time is large. Therefore, in order to always balance the amount of oxygen absorbed and desorbed by the first catalyst 32, the larger the intake air volume GA, the richer the exhaust gas flowing into the first catalyst 32 needs to be to promote oxygen release. Accordingly, in the second map, for the same maximum storage capacity Cmax, the catalyst correction value H is set to a larger value when the intake air volume GA is large.

[0031] When the first catalyst 32 deteriorates, the precious metal 32b aggregates. Consequently, the surface area of ​​the precious metal 32b decreases. When the surface area of ​​the precious metal 32b decreases, the area of ​​the precious metal 32b that oxygen reacts with when the first catalyst 32 releases oxygen decreases. In relation to this, when the first catalyst 32 deteriorates, the rate of oxygen release slows down. On the other hand, even if the first catalyst 32 deteriorates, the rate of oxygen uptake does not change. This is because oxygen uptake does not require a reaction with the precious metal 32b. For these reasons, in order to balance the amount of oxygen adsorption and desorption by the first catalyst 32, when the degree of deterioration of the first catalyst 32 increases, it is necessary to create a more favorable situation for oxygen release from the first catalyst 32. And to achieve this, the higher the degree of deterioration of the first catalyst 32, the more it is necessary to make the air-fuel ratio of the exhaust flowing into the first catalyst 32 richer. Therefore, in the second map, for the same intake air volume GA, the catalyst correction value H is set to a larger value as the maximum storage amount Cmax decreases, that is, as the degree of degradation of the first catalyst 32 increases. The catalyst correction values ​​H set in the second map are predetermined, for example, through experiments, as the optimal values ​​for balancing the oxygen absorption and desorption amounts in the first catalyst 32.

[0032] <Effects of the Embodiment> (1) Reflecting the contents of the second map described above, the CPU 82 of this embodiment calculates a large catalyst correction value H when the degree of degradation of the first catalyst 32 is high. Consequently, the target air-fuel ratio AFt and, consequently, the air-fuel ratio of the exhaust gas flowing into the first catalyst 32 are shifted to the rich side. This promotes the release of oxygen from the first catalyst 32 and allows the first catalyst 32 to exhibit high exhaust gas purification performance when the degree of degradation of the first catalyst 32 is high. With this configuration of the embodiment, the first catalyst 32 can always exhibit high exhaust gas purification performance regardless of the degree of degradation of the first catalyst 32.

[0033] (2) In this embodiment, by introducing a learned value Y, the constant error component is transferred from the feedback value FB to the learned value Y. If the feedback value FB were to include not only temporary deviations but also constant error components, it would take a long time for the feedback value FB to be updated to an appropriate value. In this respect, by transferring the constant error component to the learned value Y, as in this embodiment, the time required for the feedback value FB to be updated to an appropriate value can be shortened.

[0034] (3) In this embodiment, if the degree of degradation of the first catalyst 32 is high, the catalyst correction value H is used to intentionally shift the target air-fuel ratio AFt to the richer side. In such a situation where the target air-fuel ratio AFt is intentionally shifted to the richer side, the feedback value FB may include fluctuation components associated with the intentional enrichment, which are different from the temporary fluctuation components and the constant error components. If the learned value Y is updated under these circumstances, there is a risk that the learned value Y will include components different from the constant error components that should be included. In this case, the accuracy of air-fuel ratio control may actually worsen. That is, if fluctuation components associated with enrichment, which are different from the constant error components that should be included, are included in the learned value Y, the fluctuation components associated with enrichment will remain in the learned value Y for the period until the learned value Y returns to an appropriate value. In this case, during this period, when calculating the target air-fuel ratio AFt using the learned value Y, the fluctuation components associated with enrichment will continue to be reflected in the target air-fuel ratio AFt. In this case, if the operating conditions change during this period, for example, if the intake air volume GA decreases, then depending on those operating conditions, it may result in the target air-fuel ratio AFt reflecting fluctuating components that would not normally need to be reflected.

[0035] Therefore, in this embodiment, when the degree of degradation of the first catalyst 32 is high, the judgment value GAx that allows updating the learned value Y is set to a smaller value compared to when the degree of degradation of the first catalyst 32 is low. In other words, when the degree of degradation of the first catalyst 32 is high, updating the learned value Y is permitted only when the intake air volume GA is low, that is, only when the shift of the target air-fuel ratio AFt towards the rich side due to the catalyst correction value H is small. With such a configuration, the possibility of incorporating components other than the constant error components that are originally intended to be incorporated into the learned value Y is reduced.

[0036] <Example of changes> The above embodiment can be modified as follows. The above embodiment and the following modifications can be combined and implemented to the extent that they do not contradict each other technically.

[0037] The content of the main feedback process U5 is not limited to the examples of the above embodiment. The main feedback process U5 only needs to control the fuel injection amount of the injector 22 so that the upstream air-fuel ratio AFf matches the target air-fuel ratio AFt.

[0038] The content of the target calculation process U4 is not limited to the example of the above embodiment. The target air-fuel ratio AFt only needs to be calculated based on the feedback value FB, the learned value Y, and the catalyst correction value H. If the learning process U2 is abolished, as in the modified example described later, the target air-fuel ratio AFt only needs to be calculated based on the feedback value FB and the catalyst correction value H.

[0039] The content of the sub-feedback process U1 is not limited to the examples of the above embodiment. In the sub-feedback process U1, it is sufficient to calculate a feedback value FB to match the downstream air-fuel ratio AFr to the stoichiometric air-fuel ratio based on the difference between the downstream air-fuel ratio AFr and the stoichiometric air-fuel ratio. For example, the feedback value FB may be calculated by PI control.

[0040] The content of the learning process U2 is not limited to the example of the above embodiment. A value that reflects the constant error component related to air-fuel ratio control is called the updated value. In this process, it is sufficient to transfer such updated values ​​from the feedback value FB to the learned value Y. The updated value may be, for example, the average value of the feedback value FB over a certain period. In this process, it is sufficient to bring the feedback value FB closer to zero by updating the learned value Y according to the feedback value FB.

[0041] The contents of the first map used in the preprocessing of the learning process U2 are not limited to the examples of the above embodiment. In the first map, when the degree of degradation of the first catalyst 32 is the first degree of degradation, the judgment value GAx should be set to a smaller value compared to when the degree of degradation of the first catalyst 32 is the second degree of degradation, which is lower than the first degree of degradation. For example, in the first map, the higher the degree of degradation of the first catalyst 32, the smaller the judgment value GAx should be set to.

[0042] The learning conditions are not limited to the examples of the above embodiments. The learning conditions may include other content in place of or in addition to those of the above embodiments. Instead of defining the learning conditions by the intake air volume GA, the learning conditions may be defined by the engine load ratio. At the same time, the first map may represent the relationship between the engine load ratio, engine rotational speed, and the degree of deterioration of the first catalyst 32. In this case as well, as in the above embodiments, it is sufficient to update the learning value Y only when the shift of the target air-fuel ratio AFt towards the rich side due to the catalyst correction value H is small. The engine load ratio is the value obtained by dividing the amount of air flowing into one cylinder 12 in one cycle of the internal combustion engine 10 by the reference air volume. The reference air volume is variably set according to the engine rotational speed. The engine rotational speed is the rotational speed of the crankshaft.

[0043] The learning conditions do not necessarily have to be set to capture a situation in which the target air-fuel ratio AFt shifts slightly towards the rich side due to the catalyst correction value H. Here, the learning process U2 updates the learned value Y so that the feedback value FB can be brought closer to zero. The learning conditions only need to be set in a way that captures a situation suitable for performing this learning process U2.

[0044] The learning conditions and learning process U2 may be abolished. The content of the second map used in the characteristic processing U3 is not limited to the example of the above embodiment. The second map may be, for example, as follows: The relationship between the intake air volume GA and the catalyst correction value H is called the characteristic relationship. The second map may define only two characteristics: a first characteristic relationship that applies when the maximum absorption amount Cmax of the first catalyst 32 is below a predetermined threshold, and a second characteristic relationship that applies when the maximum absorption amount Cmax is greater than the threshold. Even in this case, when these two characteristic relationships are compared with the same intake air volume GA, it is sufficient that the catalyst correction value H of the first characteristic relationship is greater than the catalyst correction value H of the second characteristic relationship. In general, the second map should satisfy the following two conditions: The first condition is that, with respect to the same intake air volume GA, when the degree of degradation of the first catalyst 32 is the first degree of degradation, the catalyst correction value H is larger than when the degree of degradation of the first catalyst 32 is the second degree of degradation, which is lower than the first degree of degradation. The second point is that, for the same maximum storage capacity Cmax, the catalyst correction value H is larger when the intake air volume GA is the first intake volume compared to when the intake air volume GA is less than the first intake volume (second intake volume). Note that, as with the first map, the catalyst correction value H may be determined using the engine load rate instead of the intake air volume GA in the second map.

[0045] The content of characteristic processing U3 may change due to the changes in the second map as described above. In characteristic processing U3, it is sufficient to calculate a catalyst correction value H that reflects the content of the second map. • The first and second maps may be graphs or tables, or they may be mathematical formulas.

[0046] The contents of the storage amount calculation process V1 and the related specific control V are not limited to the examples of the above embodiment. The storage amount calculation process V1 and the specific control V only need to be able to appropriately calculate the maximum storage amount Cmax. Furthermore, the execution timing and frequency of the storage amount calculation process V1 and the specific control V are not limited to the examples of the above embodiment.

[0047] It is not mandatory to use the maximum absorption capacity Cmax as an indicator of the degree of degradation of the first catalyst 32. Any parameter that is appropriate as an indicator of the degree of degradation of the first catalyst 32 may be used. The content of the process for calculating the degree of degradation of the first catalyst 32 may be appropriately modified according to the parameter used.

[0048] The downstream sensor 63 may be a well-known concentration cell type oxygen sensor, that is, a sensor that can only detect whether the downstream air-fuel ratio AFr is lean or rich. The injector 22 may be of the type that directly injects fuel into the cylinder 12. [Explanation of Symbols]

[0049] 10...Internal combustion engine 12...Cylinder 22...Injector 30...Exhaust passage 32...First catalytic converter 61...Air flow meter 62...Upstream sensor 63...Downstream sensor 80...Control unit

Claims

1. This invention is applied to an internal combustion engine comprising a three-way catalytic converter located in the exhaust passage, an upstream sensor for detecting the upstream air-fuel ratio, which is the air-fuel ratio of the exhaust gas upstream of the three-way catalytic converter in the exhaust passage, a downstream sensor for detecting the downstream air-fuel ratio, which is the air-fuel ratio of the exhaust gas downstream of the three-way catalytic converter in the exhaust passage, an air flow meter for detecting the intake air volume, and an injector for injecting fuel to be supplied into the cylinder. A process to calculate a feedback value to make the downstream air-fuel ratio match the stoichiometric air-fuel ratio based on the difference between the downstream air-fuel ratio and the stoichiometric air-fuel ratio, A process for calculating a catalyst correction value such that the target air-fuel ratio in the cylinder is corrected to the rich side as the amount of intake air increases, A process for calculating the target air-fuel ratio based on the feedback value and the catalyst correction value, A process to control the fuel injection amount of the injector so that the upstream air-fuel ratio matches the target air-fuel ratio, The process of calculating the degree of degradation of the three-way catalyst can be performed. Assuming the intake air volume is the same, the process for calculating the catalyst correction value is such that the higher the degree of deterioration of the three-way catalyst, the greater the correction amount toward the rich side. Control device for internal combustion engines.

2. When predetermined learning conditions are met, it is possible to further perform a process of updating the learning value, which is used to bring the feedback value closer to zero, according to the feedback value. In the process for calculating the target air-fuel ratio, the target air-fuel ratio is calculated based on the learned value in addition to the feedback value and the catalyst correction value. A control device for an internal combustion engine according to claim 1.

Images