JP4807359B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

Description

  The present invention relates to an upstream air-fuel ratio sensor and a downstream air-space sensor disposed on the upstream side and the downstream side of a catalyst (exhaust purification catalyst, three-way catalyst) provided in the exhaust passage of the internal combustion engine. The present invention relates to an air-fuel ratio control apparatus for an internal combustion engine that controls an air-fuel ratio of an air-fuel mixture supplied to an engine based on an output value of a fuel ratio sensor (hereinafter sometimes simply referred to as “engine air-fuel ratio”).

  Conventionally, an upstream air-fuel ratio sensor, a catalyst, and a downstream air-fuel ratio sensor are provided from upstream to downstream of the exhaust passage of the internal combustion engine, and the output value of the upstream air-fuel ratio sensor and the output value of the downstream air-fuel ratio sensor are An air-fuel ratio control device that controls the air-fuel ratio of an engine based on the above is widely used.

  More specifically, the conventional air-fuel ratio control apparatus calculates a sub-feedback amount for making the output value of the downstream air-fuel ratio sensor coincide with the downstream target value. Further, the conventional air-fuel ratio control apparatus performs “learning” for updating the learning value based on “a value corresponding to the steady component included in the calculated sub feedback amount” and “updates the sub feedback amount”. Learning control to be corrected in accordance with “the learned value”. Then, the conventional air-fuel ratio control device calculates a main feedback amount based on the output value of the upstream air-fuel ratio sensor, the sub-feedback amount, and the learned value, and based on the calculated main feedback amount, the engine The air-fuel ratio (for example, fuel injection amount) is feedback controlled.

  In the present specification, the calculation of the main feedback amount is newly calculated (updated) and the use of the main feedback amount for the control of the air-fuel ratio of the engine is also referred to as executing the main feedback control. Similarly, sub-feedback control is performed by newly calculating (updating) a sub-feedback amount and using the sub-feedback amount for controlling the air-fuel ratio of the engine.

  The learning value is a value corresponding to the steady component of the sub feedback amount. The learned value is stored in a backup RAM (standby RAM) provided in the air-fuel ratio control device. The backup RAM is supplied with power from the battery regardless of the position of the ignition key switch of the vehicle on which the engine is mounted. The backup RAM can hold the “stored value (data)” as long as power is supplied from the battery. Therefore, for example, the deviation of the sub feedback amount from the steady value from the time when the sub feedback control starts when the downstream air-fuel ratio sensor is activated until the time when the sub feedback amount reaches near the steady value. It can be compensated by the learning value. As a result, immediately after the start of the sub-feedback control, the engine air-fuel ratio can be controlled to be an air-fuel ratio in the vicinity of the appropriate value. Note that the learning value may be stored in a nonvolatile memory such as an EEPROM.

  However, for example, when the power supply from the battery to the backup RAM is stopped when the battery is removed from the vehicle or when the battery is discharged, the learning value stored in the backup RAM disappears (destroyed). ) Further, the learning value in the backup RAM or the nonvolatile memory may be destroyed due to some electric noise or the like. In such a case, since the learning value is returned to the initial value (default value), it is preferable that the learning value is brought close to the appropriate value early (that is, learning is completed early).

  On the other hand, in the conventional air-fuel ratio control device, during the sub-feedback control, the “number of times of inversion” that is “the number of times the output value of the downstream air-fuel ratio sensor changes so as to cross the value corresponding to the theoretical air-fuel ratio” is a predetermined threshold value. It is determined whether or not the learning has been completed when it is determined whether or not the number of times has been exceeded, and when it has been determined that the number of inversions has exceeded the threshold number.

  That is, in the conventional air-fuel ratio control device, the output value of the downstream air-fuel ratio sensor is from a value corresponding to the air-fuel ratio leaner than the downstream target air-fuel ratio (for example, the theoretical air-fuel ratio) corresponding to the downstream target value. The number of times of changing to a value corresponding to the air-fuel ratio richer than the downstream target air-fuel ratio (hereinafter also referred to as “the number of times of rich inversion”), and the output value of the downstream air-fuel ratio sensor are the downstream target air-fuel ratio. The sum of the number of changes from the value corresponding to the air-fuel ratio richer than the fuel ratio to the value corresponding to the air-fuel ratio leaner than the downstream target air-fuel ratio (hereinafter also referred to as “the number of lean reversals”). Is counted as the number of inversions, and when the number of inversions exceeds the threshold number, it is determined that learning is completed. This is because the number of inversions is a number corresponding to the number of updates of the learning value, and therefore indicates the degree to which the learning value approaches the appropriate value, that is, the degree of convergence of the learning value.

From the above, when it is determined that the learning is not completed, the conventional air-fuel ratio control apparatus increases the opportunities for executing the sub-feedback control as much as possible, thereby increasing the learning opportunities (in other words, Then, for example, fuel cut control for stopping fuel injection is prohibited (see Patent Document 1) so that the number of inversions reaches the threshold number within a short period of time.
JP 2007-162626 A

  By the way, for the purpose of further reducing the emission amount of nitrogen oxide (NOx) in consideration of the purification performance of the catalyst, the downstream target value used for the sub-feedback control is an air-fuel ratio richer than the stoichiometric air-fuel ratio. May be set to a value corresponding to the target rich air-fuel ratio. This target rich air-fuel ratio is an air-fuel ratio within a so-called “window” range of the catalyst and is slightly richer than the stoichiometric air-fuel ratio. The target rich air-fuel ratio is an air-fuel ratio different from the stoichiometric air-fuel ratio, and is also referred to as “target non-stoichiometric air-fuel ratio” in this specification.

  As described above, when the downstream target air-fuel value is set to a value corresponding to the target rich air-fuel ratio, the temporal average value of the gas flowing into the catalyst is controlled to be slightly richer than the stoichiometric air-fuel ratio. The For this reason, since the number of inversions decreases, the time until learning is completed becomes longer. As a result, the chance that the sub feedback amount is not properly compensated by the learning value is increased, and further, the chance of fuel cut control is reduced, so that the emission may be deteriorated.

  Further, for the purpose of reducing the exhaust purification characteristics of the catalyst and the exhaust odor of the gas discharged from the engine, etc., the downstream target value used for the sub-feedback control is an air-fuel ratio leaner than the stoichiometric air-fuel ratio. The value may be set to a value corresponding to a certain target lean air-fuel ratio. This target lean air-fuel ratio is also an air-fuel ratio different from the stoichiometric air-fuel ratio, and is also referred to as “target non-stoichiometric air-fuel ratio” in this specification. Thus, when the downstream target air-fuel value is set to a value corresponding to the target lean air-fuel ratio, the number of inversions is the same as when the downstream target air-fuel ratio is set to a value corresponding to the target rich air-fuel ratio. Decreases, and the time until learning is completed becomes longer. As a result, the chance that the sub feedback amount is not properly compensated by the learning value is increased, and further, the chance of fuel cut control is reduced, so that the emission may be deteriorated.

  The present invention has been made to address the above problems. An object of the present invention is an air-fuel ratio control apparatus for an internal combustion engine that performs sub-feedback control by setting a downstream target value to a value corresponding to a “target non-stoichiometric air-fuel ratio” such as a target rich air-fuel ratio and a target lean air-fuel ratio. Therefore, an object of the present invention is to provide an air-fuel ratio control device that can complete the learning of the sub-feedback amount in a short period of time and thereby can further improve the emission.

Specifically, the air-fuel ratio control apparatus for an internal combustion engine according to the present invention is:
A catalyst provided in the exhaust passage of the internal combustion engine and having an oxygen storage function for storing oxygen ;
An upstream air-fuel ratio sensor that is disposed in a portion upstream of the catalyst in the exhaust passage and outputs an output value corresponding to the air-fuel ratio of the gas flowing through the disposed portion;
A downstream air-fuel ratio sensor that is disposed in a portion downstream of the catalyst in the exhaust passage and outputs an output value corresponding to the air-fuel ratio of the gas flowing through the disposed portion;
Sub-feedback amount calculating means for calculating a sub-feedback amount for matching the output value of the downstream air-fuel ratio sensor with the downstream target value that is the target value;
Is provided.

Furthermore, this air-fuel ratio control device
Learning to update the learning value of the sub-feedback amount based on the value corresponding to the steady component included in the sub-feedback amount every time the output value of the downstream air-fuel ratio sensor changes so as to cross the downstream target value. Learning means for performing and correcting the sub-feedback amount according to the updated learning value (change in learning value, etc.);
Learning completion determination means for acquiring a learning completion index value representing a degree of convergence of the learning value and determining whether the learning is completed based on the learning completion index value;
The internal combustion engine is configured so that the air-fuel ratio obtained based on the output value of the upstream-side air-fuel ratio sensor, the corrected sub-feedback amount, and the updated learning value matches the upstream-side target air-fuel ratio that is the target value. Air-fuel ratio control means for performing air-fuel ratio feedback control for controlling the air-fuel ratio of the air-fuel mixture supplied to the engine ;
Is provided.

In addition, this air-fuel ratio control device
The downstream target value is set to a value corresponding to the theoretical air-fuel ratio in a period before the learning completion determination unit determines that the learning is completed, and the learning completion determination unit determines that the learning is completed. And a downstream target value setting means for setting the downstream target value to a value corresponding to a target non-stoichiometric air / fuel ratio that is an air / fuel ratio other than the stoichiometric air / fuel ratio. The target non-stoichiometric air / fuel ratio is either a target rich air / fuel ratio that is richer than the stoichiometric air / fuel ratio or a target lean air / fuel ratio that is leaner than the stoichiometric air / fuel ratio.

  According to this, when the learning value of the sub-feedback amount has changed relatively greatly and the degree of convergence of the learning value is small, that is, a period before the learning completion determining unit determines that the learning is completed, The downstream target value is set to a value corresponding to the theoretical air-fuel ratio. Therefore, since the temporal average value of the air-fuel ratio of the engine is substantially the stoichiometric air-fuel ratio, the air-fuel ratio of the gas flowing out from the upstream catalyst oscillates around the stoichiometric air-fuel ratio. As a result, the “frequency with which lean inversion and rich inversion occur” increases more than when the downstream target value is set to a value corresponding to the target non-stoichiometric air-fuel ratio. That is, the learning value update opportunity increases as the number of inversions increases early, so that learning can be completed early.

in this case,
The learning completion judging means
The number of inversions, which is the number of times the output value of the downstream air-fuel ratio sensor changes so as to cross a value corresponding to the theoretical air-fuel ratio, is acquired as the learning completion index value, and the acquired number of inversions is equal to or greater than a predetermined threshold number of times. It is preferable to be configured to determine that the learning is completed when it is determined that the learning has been completed.

  The number of inversions increases in accordance with the number of learning value updates. Therefore, according to the above configuration, the learning completion index value indicating the degree of convergence of the learning value can be obtained accurately and easily.

Further, the sub-feedback amount calculating means includes
It is preferable that the learning value is corrected by an amount corresponding to a difference between the theoretical air-fuel ratio and the target non-stoichiometric air-fuel ratio when the learning completion determining unit determines that the learning is completed. .

  The learning value obtained before the learning completion determination is a value based on the sub-feedback amount calculated under the state where the downstream target value is set to a value corresponding to the theoretical air-fuel ratio. On the other hand, after the learning completion determination, the sub-feedback control is executed in a state where the downstream target value is set to a value corresponding to “the target non-stoichiometric air / fuel ratio that is either the target rich air / fuel ratio or the target lean air / fuel ratio”. Is done. Therefore, the learning value itself obtained before the learning completion determination is not appropriate as the “learning value after the learning completion determination”. Therefore, as described above, when the learning completion determination unit determines that the learning is completed, the learning value is corrected by an amount corresponding to the difference between the theoretical air-fuel ratio and the target non-stoichiometric air-fuel ratio. According to this, the learning value obtained before the learning completion determination can be converted into a learning value having an appropriate value after the learning completion determination.

(Preferred embodiment)
An air-fuel ratio control apparatus for this internal combustion engine,
A first condition is set as a sub-feedback control condition before it is determined that the learning is completed, and a second condition in which a further condition is added to the first condition is determined after the learning is determined to be completed. With sub feedback control condition setting means for setting as the sub feedback control condition,
The sub feedback amount calculating means is configured to calculate the sub feedback amount only when the sub feedback control condition is satisfied,
It is preferable that the learning unit is configured to perform “correction according to the learning value of the learning and the sub feedback amount” only when the sub feedback control condition is satisfied.

  According to this, until the learning is completed, the sub feedback control condition (first condition) that is relaxed than after the learning is completed is set, so that the sub feedback control is easily performed. As a result, learning opportunities increase, and the learning completion time can be advanced.

Furthermore, the air-fuel ratio control apparatus for the internal combustion engine is
The fuel cut control for stopping the fuel supply to the engine is not executed even if a predetermined fuel cut condition is satisfied in a period before the learning completion determining means determines that the learning is completed. It is preferable that fuel cut control means for executing the fuel cut control when the fuel cut condition is satisfied in the period after the learning completion determination means determines that the learning is completed is preferable.

  According to this, since the fuel cut control is not executed until learning is completed, opportunities for learning can be increased. Further, after the learning is completed, fuel cut control is executed, so that it is possible to protect the catalyst and improve the emission. In this case, the sub-feedback control condition includes not being in fuel cut control.

  Embodiments of an air-fuel ratio control device for an internal combustion engine (hereinafter simply referred to as “control device”) according to the present invention will be described below with reference to the drawings. This control device is also a fuel injection amount control device that controls the fuel injection amount in order to control the air-fuel ratio of the internal combustion engine.

(Constitution)
FIG. 1 shows a schematic configuration of a system in which this control device is applied to a four-cycle spark ignition multi-cylinder (four-cylinder) internal combustion engine 10. FIG. 1 shows only a cross section of a specific cylinder, but the other cylinders have the same configuration.

  The internal combustion engine 10 includes a cylinder block portion 20 including a cylinder block, a cylinder block lower case, an oil pan, and the like, a cylinder head portion 30 fixed on the cylinder block portion 20, and a gasoline mixture to the cylinder block portion 20. An intake system 40 for supplying and an exhaust system 50 for releasing exhaust gas from the cylinder block unit 20 to the outside are included.

  The cylinder block unit 20 includes a cylinder 21, a piston 22, a connecting rod 23, and a crankshaft 24. The piston 22 reciprocates in the cylinder 21, and the reciprocating motion of the piston 22 is transmitted to the crankshaft 24 through the connecting rod 23, whereby the crankshaft 24 rotates. The wall surface of the cylinder 21 and the upper surface of the piston 22 form a combustion chamber 25 together with the lower surface of the cylinder head portion 30.

  The cylinder head portion 30 includes an intake port 31 communicating with the combustion chamber 25, an intake valve 32 that opens and closes the intake port 31, an intake camshaft that drives the intake valve 32, and a phase angle and lift amount of the intake camshaft are continuously provided. Variable intake timing device 33 to be changed, actuator 33a of variable intake timing device 33, exhaust port 34 communicating with combustion chamber 25, exhaust valve 35 for opening and closing exhaust port 34, exhaust camshaft 36 for driving exhaust valve 35, An ignition plug 37, an igniter 38 including an ignition coil that generates a high voltage to be applied to the ignition plug 37, and an injector (fuel injection means) 39 for injecting fuel into the intake port 31 are provided.

  The intake system 40 is provided in an intake pipe 41 including an intake manifold that communicates with the intake port 31 and forms an intake passage together with the intake port 31, an air filter 42 provided at an end of the intake pipe 41, and the intake pipe 41. A throttle valve 43 is provided which makes the opening cross-sectional area of the intake passage variable. The throttle valve 43 is rotationally driven in the intake pipe 41 by a throttle valve actuator 43a made of a DC motor.

  The exhaust system 50 includes an exhaust manifold 51 including a plurality of branches connected at one end to the exhaust port 34 of each cylinder, and the other ends of the branches of each exhaust manifold 51 and all branches are assembled. An exhaust pipe 52 connected to the collecting portion, an upstream catalyst 53 disposed in the exhaust pipe 52, and a downstream catalyst 54 disposed in the exhaust pipe 52 downstream of the upstream catalyst 53 are provided. The exhaust port 34, the exhaust manifold 51, and the exhaust pipe 52 constitute an exhaust passage.

Each of the upstream catalyst 53 and the downstream catalyst 54 is a so-called three-way catalyst device (exhaust gas purification catalyst) carrying an active component made of a noble metal such as platinum. Each catalyst has a function of oxidizing unburned components such as HC and CO and reducing nitrogen oxides (NOx) when the air-fuel ratio of the gas flowing into each catalyst is the stoichiometric air-fuel ratio. This function is also called a catalyst function. Furthermore, each catalyst has an oxygen storage function for storing (storing) oxygen, and even if the air-fuel ratio shifts from the stoichiometric air-fuel ratio by this oxygen storage function, unburned components and nitrogen oxides can be purified. . This oxygen storage function is provided by ceria (CeO ₂ ) supported on the catalyst.

  On the other hand, this system includes a hot-wire air flow meter 61, a throttle position sensor 62, a cam position sensor 63, a crank position sensor 64, a water temperature sensor 65, an upstream air-fuel ratio sensor 66, a downstream air-fuel ratio sensor 67, and an accelerator opening sensor 68. It has.

The air flow meter 61 outputs a signal corresponding to the mass flow rate Ga of the intake air flowing through the intake pipe 41.
The throttle position sensor 62 detects the opening (throttle valve opening) of the throttle valve 43 and outputs a signal representing the throttle valve opening TA.

The cam position sensor 63 generates a signal (G2 signal) having one pulse every time the intake camshaft rotates 90 ° (that is, every time the crankshaft 24 rotates 180 °).
The crank position sensor 64 outputs a signal having a narrow pulse every time the crankshaft 24 rotates 10 °, and a wide pulse every time the crankshaft 24 rotates 360 °. This signal is converted into an engine speed NE by an electric control device 70 described later.
The water temperature sensor 65 detects the temperature of the cooling water of the internal combustion engine 10 and outputs a signal representing the cooling water temperature THW.

  The upstream air-fuel ratio sensor 66 is disposed in the exhaust passage. The upstream air-fuel ratio sensor 66 is disposed at the downstream side of the collection portion of the branches of the exhaust manifold 51 or the collection portion thereof. The upstream air-fuel ratio sensor 66 is a limiting current type oxygen concentration sensor. As shown in FIG. 2, the upstream air-fuel ratio sensor 66 outputs an output value Vabyfs that is a voltage corresponding to the air-fuel ratio A / F of the “detected gas”. Therefore, in this example, the upstream air-fuel ratio sensor 66 is the air-fuel ratio of the gas flowing through the exhaust passage and the portion where the upstream air-fuel ratio sensor 66 is disposed (and therefore the gas flowing into the upstream catalyst 53). An output value Vabyfs corresponding to the air-fuel ratio and the air-fuel ratio of the air-fuel mixture supplied to the engine is generated.

  This output value Vabyfs matches the value Vstoich when the air-fuel ratio of the detected gas is the stoichiometric air-fuel ratio. The output value Vabyfs increases as the air-fuel ratio of the gas to be detected increases (lean). That is, the upstream air-fuel ratio sensor 66 is a wide-area air-fuel ratio sensor whose output continuously changes in response to changes in the air-fuel ratio of the gas to be detected.

  The electric control device 70 described later stores the table (map) Mapabyfs shown in FIG. 2, and detects the air-fuel ratio by applying the actual output value Vabyfs to the table Mapabyfs (acquires the detected air-fuel ratio abyfs). To do). Hereinafter, the air-fuel ratio acquired from the output value Vabyfs of the upstream air-fuel ratio sensor and the table Mapabyfs is also referred to as upstream air-fuel ratio abyfs.

  The downstream air-fuel ratio sensor 67 is an exhaust passage that is downstream of the upstream catalyst 53 and upstream of the downstream catalyst 54 (that is, an exhaust passage between the upstream catalyst 53 and the downstream catalyst 54). ). The downstream air-fuel ratio sensor 67 is a known electromotive force type oxygen concentration sensor (a known concentration cell type oxygen concentration sensor using stabilized zirconia). The downstream air-fuel ratio sensor 67 is an air-fuel ratio of a gas to be detected that is a gas that flows through a portion of the exhaust passage where the downstream air-fuel ratio sensor 67 is disposed (accordingly, an empty of the gas flowing into the downstream catalyst 54). An output value Voxs corresponding to the fuel ratio and the time average value of the air-fuel ratio of the air-fuel mixture supplied to the engine is generated.

  As shown in FIG. 3, the output value Voxs becomes the maximum output value max (for example, about 0.9 V) when the air-fuel ratio of the detected gas is richer than the stoichiometric air-fuel ratio, and the air-fuel ratio of the detected gas is When the air-fuel ratio is leaner than the stoichiometric air-fuel ratio, the minimum output value min (for example, about 0.1 V) is obtained. When the air-fuel ratio of the gas to be detected is the stoichiometric air-fuel ratio, the voltage Vst approximately halfway between the maximum output value max and the minimum output value min. (Intermediate voltage Vst, for example, about 0.5 V). Further, this output value Voxs suddenly changes from the maximum output value max to the minimum output value min when the air-fuel ratio of the gas to be detected changes from an air-fuel ratio richer than the stoichiometric air-fuel ratio to a lean air-fuel ratio. When the air-fuel ratio of the detection gas changes from an air-fuel ratio leaner than the stoichiometric air-fuel ratio to a rich air-fuel ratio, it suddenly changes from the minimum output value min to the maximum output value max.

  Referring to FIG. 1 again, the accelerator opening sensor 68 outputs a signal representing the operation amount Accp of the accelerator pedal AP operated by the driver.

  The electric control device 70 includes a CPU 71 connected to each other by a bus, a ROM 72 in which programs executed by the CPU 71, tables (maps, functions), constants, and the like are stored in advance, a RAM 73 in which the CPU 71 temporarily stores data as necessary, And a microcomputer including an interface 75 including a backup RAM 74 and an AD converter.

  The backup RAM 74 is supplied with electric power from a battery mounted on the vehicle regardless of the position of an ignition key switch (not shown) of the vehicle on which the engine 10 is mounted (any one of an off position, a start position, an on position, etc.). It is like that. When receiving power from the battery, the backup RAM 74 stores data (data is written) in accordance with an instruction from the CPU 71 and holds (stores) the data so that the data can be read. The backup RAM 74 cannot retain data when power supply from the battery is interrupted, for example, when the battery is removed from the vehicle. Therefore, when the power supply to the backup RAM 74 is resumed, the CPU 71 initializes (sets to a default value) data to be held in the backup RAM 74.

  The interface 75 is connected to the sensors 61 to 68 and supplies signals from the sensors 61 to 68 to the CPU 71. Further, the interface 75 sends drive signals (instruction signals) to the actuator 33a, the igniter 38, the injector 39, the throttle valve actuator 43a, etc. of the variable intake timing device 33 in accordance with instructions from the CPU 71.

(Control outline)
Next, an outline of the operation of the control device configured as described above will be described.
This control device includes main feedback control for matching the upstream air-fuel ratio abyfs obtained based on the output value Vabyfs of the upstream air-fuel ratio sensor 66 with the upstream target air-fuel ratio abyfr, and the output value Voxs of the downstream air-fuel ratio sensor 67. And air-fuel ratio feedback control including sub-feedback control that matches the downstream target value Voxsref. Actually, the control device sets “upstream air-fuel ratio abyfs” to “sub-feedback amount Vafsfb calculated so as to reduce the output deviation amount Dvoxs between the output value Voxs of the downstream air-fuel ratio sensor 67 and the downstream target value Voxsref. And its learning value Vafsfbg ”, thereby calculating“ feedback control air-fuel ratio (corrected detected air-fuel ratio) abyfsc ”and making the feedback control air-fuel ratio abyfsc coincide with the upstream target air-fuel ratio abyfr Take control.

<Main feedback control>
More specifically, the control device calculates the feedback control output value Vabyfc according to the following equation (1). In equation (1), Vabyfs is the output value of the upstream air-fuel ratio sensor 66, Vafsfb is a sub-feedback amount calculated based on the output value Voxs of the downstream air-fuel ratio sensor 67, and Vafsfbg is a learning value of the sub-feedback amount. . These values are all values obtained at the present time. A method of calculating the sub feedback amount Vafsfb and the learning value Vafsfbg will be described later.
Vabyfc = Vabyfs + Vafsfb + Vafsfbg (1)

As shown in the following equation (2), the control device obtains the feedback control air-fuel ratio abyfsc by applying the feedback control output value Vabyfc to the table Mapabyfs shown in FIG.
abyfsc = Mapabyfs (Vabyfc) (2)

  On the other hand, the control device obtains the in-cylinder intake air amount Mc (k) that is the amount of air sucked into the cylinder at the present time. The in-cylinder intake air amount Mc (k) is obtained on the basis of the output Ga of the air flow meter 61 and the engine rotational speed NE for each intake stroke of each cylinder (for example, with respect to the output Ga of the air flow meter 61) The value obtained by dividing the first-order lag process by the engine speed NE) is stored in the RAM 73 while corresponding to each intake stroke. The in-cylinder intake air amount Mc (k) may be calculated by a well-known air model (a model constructed according to a physical law simulating the behavior of air in the intake passage).

As shown in the following equation (3), the control device obtains the basic fuel injection amount Fbase by dividing the in-cylinder intake air amount Mc (k) by the current upstream target air-fuel ratio abyfr (k). The upstream target air-fuel ratio abyfr (k) is set to the stoichiometric air-fuel ratio except for special cases such as during engine warm-up, during fuel cut recovery, and during catalyst overheat prevention. The increase after returning from the fuel cut is a predetermined period after the fuel injection stop (fuel cut), which will be described later, is completed and the fuel injection is restarted (for example, the oxygen storage amount of the upstream catalyst 53 is the maximum oxygen storage amount of the upstream catalyst 53). This is carried out by setting the upstream target air-fuel ratio abyfr (k) to a richer air-fuel ratio than the stoichiometric air-fuel ratio during the period until it drops below about half of Cmax1). The upstream target air-fuel ratio abyfr (k) is also stored in the RAM 73 while corresponding to each intake stroke.
Fbase = Mc (k) / abyfr (k) (3)

As shown in the following equation (4), the control device corrects the basic fuel injection amount Fbase with the main feedback amount DFi (adds the main feedback amount DFi to the basic fuel injection amount Fbase) to thereby obtain the final fuel injection amount Fi. Is calculated. Then, the control device injects the fuel of the final fuel injection amount Fi from the injector 39 of the cylinder that is in the intake stroke.
Fi = Fbase + DFi (4)

The main feedback amount DFi in the above equation (4) is obtained as follows. As shown in the following equation (5), the control device feedback-controls the in-cylinder intake air amount Mc (k−N) at a time point N cycles (that is, N · 720 ° crank angle) before the current time point. By dividing by the air-fuel ratio abyfsc, “in-cylinder supplied fuel amount Fc (k−N)”, which is the amount of fuel actually supplied to the combustion chamber 25 at the time N cycles before the current time, is obtained.
Fc (k−N) = Mc (k−N) / abyfsc (5)

  Thus, in order to obtain the in-cylinder fuel supply amount Fc (k−N) N cycles before the present time, the in-cylinder intake air amount Mc (k−N) N strokes before the present time is used as the feedback control air-fuel ratio abyfsc. This is because it takes a time corresponding to the N stroke until the air-fuel mixture combusted in the combustion chamber 25 reaches the upstream air-fuel ratio sensor 66.

Next, as shown in the following equation (6), the control device calculates the in-cylinder intake air amount Mc (k−N) N strokes before the current stroke to the upstream target air-fuel ratio abyfr (k The target in-cylinder fuel supply amount Fcr (k−N) N strokes before the current stroke is obtained by dividing by −N).
Fcr (k−N) = Mc (k−N) / abyfr (k−N) (6)

As shown in the following equation (7), the control device obtains the in-cylinder fuel supply amount deviation by subtracting the in-cylinder fuel supply amount Fc (k−N) from the target in-cylinder fuel supply amount Fcr (k−N). Set as DFc. This in-cylinder fuel supply amount deviation DFc is an amount representing the excess or deficiency of the fuel supplied into the cylinder at the time point before the N stroke.
DFc = Fcr (kN) -Fc (kN) (7)

Thereafter, the control device obtains the main feedback amount DFi based on the following equation (8). In this equation (8), Gp is a preset proportional gain, and Gi is a preset integral gain. The coefficient KFB in the equation (8) is preferably variable depending on the engine speed NE, the in-cylinder intake air amount Mc, and the like, but is set to “1” here. Further, the value SDFc in the equation (8) is an integral value of the in-cylinder fuel supply amount deviation DFc. That is, the control device calculates the main feedback amount DFi by proportional-integral control based on the feedback control air-fuel ratio abyfsc and the upstream target air-fuel ratio abyfr. The main feedback amount DFi is added to the basic fuel injection amount Fbase as shown in the above equation (4), thereby calculating the final fuel injection amount Fi.
DFi = (Gp · DFc + Gi · SDFc) · KFB (8)

<Sub feedback control>
The control device calculates the sub feedback amount Vafsfb described above as follows.
That is, the control device obtains the output deviation amount DVoxs by subtracting the current output value Voxs of the downstream air-fuel ratio sensor 67 from the downstream target value Voxsref as shown in the following equation (9).
DVoxs = Voxsref−Voxs (9)

  The downstream target value Voxsref in the equation (9) is determined so that the purification efficiency of the upstream catalyst 53 is good. As described later, the downstream target value Voxsref is determined that learning of the sub feedback amount Vafsfb by learning control is not completed (hereinafter referred to as “before learning completion determination”, “before learning completion” or “learning”). And also when it is determined that learning of the sub-feedback amount Vafsfb by learning control has been completed (hereinafter also referred to as “after learning completion determination” or “after learning completion”). , Are set to different values.

  That is, the downstream target value Voxsref will be output by the downstream air-fuel ratio sensor 67 before “learning completion determination” when “the theoretical air-fuel ratio gas reaches the downstream air-fuel ratio sensor 67 constantly”. The value (the stoichiometric air-fuel ratio equivalent value Vst shown in FIG. 3) is set. On the other hand, the downstream target value Voxsref is determined by the downstream air-fuel ratio sensor 67 after “learning completion determination” within a window W of the catalyst (upstream catalyst 53 and downstream catalyst 54) (a predetermined value including the theoretical air-fuel ratio). The downstream air-fuel ratio sensor 67 outputs when the “target rich air-fuel ratio gas that is within the air-fuel ratio range) and is slightly richer than the stoichiometric air-fuel ratio is steadily reached. The wax value is set to the target rich air-fuel ratio equivalent value Vrich shown in FIG.

The control device obtains the sub feedback amount Vafsfb based on the following equation (10). In equation (10), Kp is a preset proportional gain (proportional constant), and Ki is a preset integral gain (integral constant). SDVoxs is an integral value of the output deviation amount DVoxs. As will be described later, the integral gain Ki is set to a relatively large first integral gain KiLarge before the learning completion determination, and is set to a second integral gain KiSmall smaller than the first integral gain after the learning completion determination (FIG. 4). (See (D) of).
Vafsfb = Kp · DVoxs + Ki · SDVoxs (10)

  As described above, the control device calculates the sub feedback amount Vafsfb by proportional-integral control based on the output value Voxs of the downstream air-fuel ratio sensor 67 and the downstream target value Voxsref. The sub feedback amount Vafsfb is used to calculate the feedback control output value Vabyfc, as shown in the above-described equation (1).

<Learning sub-feedback control>
The control device updates the learning value Vafsfbg of the sub feedback amount Vafsfb based on the following equation (11) before the learning completion determination. The learning value Vafsfbg is updated at the time of rich inversion and lean inversion. Note that the left side Vafsfbgnew in the equation (11) represents the updated learning value Vafsbfbg.

As is clear from the equation (11), the learning value Vafsfbg is a value obtained by applying a filter process for noise removal to the integral term Ki · SDVoxs of the sub feedback amount Vafsfb. In the formula (11), the value α is an arbitrary value of 0 or more and less than 1. By setting the value α to “0”, the integral term Ki · SDVoxs may be taken in as it is as the learning value Vafsfbg. The learning value Vafsfbg is stored in the backup RAM 74.
Vafsfbgnew = α · Vafsfbg + (1-α) · Ki · SDVoxs (11)

  As described above, lean inversion means that the output value Voxs of the downstream air-fuel ratio sensor 67 changes from a value larger than the downstream target value Voxsref to a value smaller than the downstream target value Voxsref. The rich inversion means that the output value Voxs of the downstream air-fuel ratio sensor 67 changes from a value smaller than the downstream target value Voxsref to a value larger than the downstream target value Voxsref.

<Correction of sub-feedback amount accompanying learning of sub-feedback control>
As shown in the above equation (1), the control device obtains the feedback control output value Vabyfc by adding the sub feedback amount Vafsfb and the learned value Vafsfbg to the output value Vabyfs of the upstream air-fuel ratio sensor 66. The learning value Vafsfbg is a value obtained by incorporating all or part of the integral term Ki · SDVoxs (stationary component) of the sub feedback amount Vafsfb. Therefore, when the learning value Vafsfbg is updated, if the sub feedback amount Vafsfb is not corrected according to the updated amount, double correction is performed by the updated learning value Vafsfbg and the sub feedback amount Vafsfb. Therefore, when the learning value Vafsfbg is updated, it is necessary to correct the sub feedback amount Vafsfb according to the updated amount of the learning value Vafsbfbg.

Therefore, as shown in the following equations (12) and (13), when the learning value Vafsfbg is updated so as to increase by the change amount ΔG, the control device decreases the sub feedback amount Vafsfb by the change amount ΔG. In the equation (12), Vafsfbg0 is the learning value Vafsfbg immediately before the update. Accordingly, the change amount ΔG is a positive value or a negative value.
ΔG = Vafsfbg−Vafsfbg0 (12)
Vafsfb = Vafsfb−ΔG (13)

  As described above, the control device corrects the output value Vabyfs of the upstream air-fuel ratio sensor 66 by the sum of the sub-feedback amount Vafsfb and the learning value Vafsfbg, and obtains the feedback control output value Vabyfc obtained by the correction. Based on this, the feedback control air-fuel ratio abyfsc is acquired. Then, the control device controls the fuel injection amount Fi so that the acquired feedback control air-fuel ratio abyfsc matches the upstream target air-fuel ratio abyfr. As a result, the upstream air-fuel ratio abyfs corresponding to the output value Vabyfs of the upstream air-fuel ratio sensor 66 approaches the upstream target air-fuel ratio abyfr, and at the same time, the output value Voxs of the downstream air-fuel ratio sensor 67 becomes the downstream target value Voxsref. Get closer. That is, the control device is an air-fuel ratio feedback control means for matching the air-fuel ratio of the engine mixture to the upstream target air-fuel ratio abyfr based on the output value Vabyfs of the upstream air-fuel ratio sensor 66, the sub-feedback amount Vafsfb, and the learned value Vafsfbg. It has.

<Sub feedback feedback learning completion determination>
As described above, the control device updates the learning value Vafsfbg for each rich inversion and each lean inversion. Therefore, the sum of the rich inversion number and the lean inversion number (also simply referred to as “inversion number C”) is a value indicating how close the learning value Vafsfbg is to the appropriate value, that is, the degree of convergence of the learning value Vafsfbg. Is a learning completion index value representing

  Therefore, as shown in the time chart of FIG. 4, the control device counts the number of inversions C during the sub-feedback control before the learning completion determination, and whether or not the counted number of inversions C is equal to or greater than the threshold number of times Cth. Determine whether. When the controller determines that the number of inversions C is equal to or greater than the threshold number of times Cth, the control device determines that learning has been completed (the learning value Vafsfbg has reached an appropriate value) (see time t1 in FIG. 4). At this time, the control device changes the value of the learning completion flag XGK from “0” to “1”. That is, the learning completion flag XGK indicates that the learning is completed when the value is “0”, and after the learning is completed when the value is “1”. Indicates that there is. When power supply to the backup RAM 74 is resumed, that is, when data including the learning value Vafsfbg to be held in the backup RAM 74 is initialized, the value of the learning completion flag XGK is set to “0”. .

<Process associated with learning completion determination of sub feedback control>
<< Change of downstream target value >>
As shown in FIG. 4, before the learning completion determination (before time t1 in FIG. 4), the control device sets the downstream target value Voxsref to a value Vst corresponding to the theoretical air-fuel ratio. As a result, since the temporal average value of the air-fuel ratio of the engine becomes substantially the stoichiometric air-fuel ratio, the air-fuel ratio of the gas flowing out from the upstream side catalyst 53 oscillates around the stoichiometric air-fuel ratio. Therefore, “the frequency with which lean inversion and rich inversion occur” increases more than when the downstream target value Voxsref is set to a value Vrich corresponding to the target rich air-fuel ratio. That is, the number of inversions increases early, and the learning value Vafsfbg is updated quickly.

  By the way, the “purification rate of unburned matter” of the three-way catalyst such as the upstream catalyst 53 and the downstream catalyst 54 decreases relatively slowly even if the air-fuel ratio of the engine shifts slightly to the rich side from the window W. On the other hand, the “nitrogen oxide purification rate” of the three-way catalyst rapidly decreases when the air-fuel ratio of the engine shifts slightly leaner than the window W. Accordingly, after the learning completion determination (after time t1 in FIG. 4), the control device changes the downstream target value Voxsref to a value Vrich corresponding to the “target rich air-fuel ratio” that is an air-fuel ratio richer than the stoichiometric air-fuel ratio. To do. As a result, the temporal average value of the air-fuel ratio of the engine becomes slightly richer than the stoichiometric air-fuel ratio (target rich air-fuel ratio). Therefore, even if the air-fuel ratio of the engine is displaced to the lean side, the nitrogen oxide purification rate of the three-way catalyst is difficult to decrease, so that it is possible to avoid an increase in the emission amount of nitrogen oxides. However, depending on the three-way catalyst, setting the downstream target value Voxsref to a value corresponding to the “target lean air-fuel ratio”, which is an air-fuel ratio leaner than the stoichiometric air-fuel ratio, causes exhaust purification performance and / or exhaust odor, etc. It may be advantageous from a viewpoint.

<< Correction of learning value Vafsfbg >>
Further, at the time of learning completion determination (time t1 in FIG. 4), the control device sets the learning value Vafsfbg between the theoretical air-fuel ratio and the target rich air-fuel ratio (target non-theoretical air-fuel ratio) as shown in the following equation (14). It is corrected (increased) by an amount β corresponding to the difference.
Vafsfb = Vafsfb + β (14)

  This is because before the learning completion determination, the sub-feedback control is executed in a state where the downstream target value Voxsref is set to a value Vst corresponding to the theoretical air-fuel ratio, and the learning value Vafsfbg is acquired in that state, while learning After the completion determination, the sub-feedback control is executed in a state where the downstream target value Voxsref is set to a value Vrich corresponding to the target rich air-fuel ratio, and the learning value Vafsfbg is used for determining the fuel injection amount Fi. It is. That is, at the time of learning completion determination, the control device adds the correction amount β to the learning value Vafsfbg obtained up to that time, so that the learning value Vafsfbg obtained up to that time is appropriate for air-fuel ratio control after that time. To the learning value.

<< Change of integral gain >>
In addition, before the learning completion determination (before time t1 in FIG. 4), the control device has a relatively large value (first integral gain KiLarge) for the integral gain Ki when obtaining the integral term Ki · SDVoxs of the sub feedback amount Vafsfb. Is set. As a result, the air-fuel ratio of the engine is likely to vibrate, so that the frequency of occurrence of lean inversion and rich inversion increases. That is, the number of inversions increases early, and the learning value Vafsfbg is updated quickly, so that the learning completion time can be advanced.

  Furthermore, after the learning completion determination (after time t1 in FIG. 4), the control device sets the integral gain Ki to a second integral gain KiSmall that is smaller than the first integral gain KiLarge. Thereby, the fluctuation of the air-fuel ratio of the engine becomes gentle after the learning completion determination, so that the updated learning value Vafsfbg becomes more stable. As a result, emissions can be further improved.

(Actual operation)
Next, actual operation of the air-fuel ratio control apparatus configured as described above will be described.

<Fuel injection amount control>
The CPU 71 performs the routine for calculating the final fuel injection amount Fi and instructing the fuel injection shown in FIG. 5, and the crank angle of a predetermined cylinder becomes a predetermined crank angle before the intake top dead center (for example, BTDC 90 ° CA). Each time, it is repeatedly executed for that cylinder (hereinafter also referred to as “fuel injection cylinder”). Therefore, when the predetermined timing is reached, the CPU 71 starts the process from step 500 and proceeds to step 510 to determine whether or not the fuel cut control condition is satisfied. The fact that the fuel cut control condition is satisfied means that the fuel cut return condition is not satisfied after the fuel cut start condition described below is satisfied.

The fuel cut start condition is satisfied only when both condition 1 and condition 2 described below are satisfied.
(Condition 1) The throttle valve opening TA is “0 (or a predetermined opening or less)”. That is, the throttle valve 43 is fully closed. The CPU 71 controls the opening degree of the throttle valve 43 so as to increase as the accelerator pedal operation amount Accp increases.
(Condition 2) The engine speed NE is equal to or higher than the fuel cut speed NEFC.

The fuel cut return condition is satisfied when either of the conditions 3 and 4 described below is satisfied, and is not satisfied when both of the conditions 3 and 4 are not satisfied.
(Condition 3) The throttle valve opening TA is larger than “0 (the predetermined opening)”.
(Condition 4) The engine rotational speed NE is smaller than the fuel cut return rotational speed NEFK (NEFK = NEFC−ΔN) which is smaller than the fuel cut rotational speed NEFC by a predetermined rotational speed ΔN.

  Assume that the fuel cut control condition is not satisfied. In this case, the CPU 71 makes a “No” determination at step 510, sequentially performs the processing from step 520 to step 550 described below, proceeds to step 595, and ends this routine once.

Step 520: The CPU 71 determines an in-cylinder intake air amount Mc (k), which is the amount of air taken into the fuel injection cylinder, based on the intake air amount Ga measured by the air flow meter 61, the engine speed NE, and the map f. get.
Step 530: The CPU 71 obtains the basic fuel injection amount Fbase according to the above equation (3).
Step 540: The CPU 71 obtains the final fuel injection amount Fi according to the above equation (4).
Step 550: The CPU 71 injects the fuel of the final fuel injection amount Fi from the injector 39 provided corresponding to the fuel injection cylinder.

  Thus, the fuel of the final fuel injection amount Fi, which is the basic fuel injection amount Fbase that is feedback-corrected by the main feedback amount DFi, is injected into the fuel injection cylinder.

  On the other hand, if the fuel cut control condition is satisfied, the CPU 71 determines “Yes” in step 510 and proceeds to step 560 to determine whether or not the value of the learning completion flag XGK is “1”.

  Now, if the learning completion flag XGK is “0” before the learning completion determination, the CPU 71 determines “No” in step 560 and sequentially performs the processing from step 520 to step 550 described above. Proceeding to step 595, the present routine is temporarily terminated. As a result, since the process of step 550 is executed, even if the fuel cut control condition is satisfied, the fuel cut (stop of fuel injection) is not executed if it is before the learning completion determination.

  On the other hand, if the CPU 71 proceeds to step 560 after the learning completion determination and the value of the learning completion flag XGK is “1”, the CPU 71 determines “Yes” in step 560 and then proceeds to step 595. Proceed directly to end this routine. As a result, since the process of step 550 is not executed, fuel cut is executed.

<Calculation of main feedback amount>
The CPU 71 repeatedly executes the main feedback amount calculation routine shown in the flowchart of FIG. 6 every elapse of a predetermined time. Therefore, when the predetermined timing is reached, the CPU 71 starts the process from step 600 and proceeds to step 605 to determine whether or not the main feedback control condition (upstream air-fuel ratio feedback control condition) is satisfied. The main feedback control condition is, for example, that fuel cut is not being performed, the engine coolant temperature THW is equal to or higher than a first predetermined temperature, the intake air amount (load) per rotation of the engine is equal to or lower than a predetermined value, and upstream This is established when the side air-fuel ratio sensor 66 is activated.

  Now, if the description is continued assuming that the main feedback control condition is satisfied, the CPU 71 determines “Yes” in step 605, performs the processing of step 610 to step 640 described below in order, and proceeds to step 695. This routine is temporarily terminated.

Step 610: The CPU 71 acquires the feedback control output value Vabyfc according to the above equation (1).
Step 615: The CPU 71 obtains the feedback control air-fuel ratio abyfsc according to the above equation (2).
Step 620: The CPU 71 acquires the in-cylinder fuel supply amount Fc (k−N) according to the above equation (5).
Step 625: The CPU 71 acquires the target in-cylinder fuel supply amount Fcr (k−N) according to the above equation (6).

Step 630: The CPU 71 acquires the in-cylinder fuel supply amount deviation DFc according to the above equation (7).
Step 635: The CPU 71 acquires the main feedback amount DFi according to the above equation (8). In this example, the coefficient KFB is set to “1”. The integrated value SDFc of the in-cylinder fuel supply amount deviation DFc is obtained in the next step 640.
Step 640: The CPU 71 adds the in-cylinder fuel supply amount deviation DFc obtained in the above step 630 to the integral value SDFc of the in-cylinder fuel supply amount deviation DFc at that time, so that a new in-cylinder fuel supply amount deviation is obtained. An integral value SDFc is obtained.

  Thus, the main feedback amount DFi is obtained by the proportional integral control, and this main feedback amount DFi is reflected in the final fuel injection amount Fi by step 540 of FIG. As a result, since the excess or deficiency of the fuel supply amount before the N stroke from the present time is compensated, the average value of the air / fuel ratio of the engine (therefore, the air / fuel ratio of the gas flowing into the upstream catalyst 53) becomes the upstream target air / fuel ratio abyfr. (Stoichiometric air-fuel ratio except for special cases)

  On the other hand, if the main feedback control condition is not satisfied at the time of determination in step 605, the CPU 71 determines “No” in step 605 and proceeds to step 645 to set the value of the main feedback amount DFi to “0”. To do. Next, in step 650, the CPU 71 stores “0” in the integral value SDFc of the in-cylinder fuel supply amount deviation. Thereafter, the CPU 71 proceeds to step 695 to end the present routine tentatively. Thus, when the main feedback control condition is not satisfied, the main feedback amount DFi is set to “0”. Accordingly, the basic fuel injection amount Fbase is not corrected by the main feedback amount DFi.

<Calculation of sub feedback amount and learning value>
The CPU 71 executes the routine shown in FIG. 7 every elapse of a predetermined time in order to calculate the sub feedback amount Vafsfb and the learning value Vafsfbg of the sub feedback amount Vafsfb. Accordingly, when the predetermined timing is reached, the CPU 71 starts processing from step 700 and proceeds to step 705 to determine whether or not the sub feedback control condition is satisfied. As the sub feedback control condition, for example, the main feedback control condition in step 605 described above is satisfied, the upstream target air-fuel ratio abyfr is set to the stoichiometric air-fuel ratio, and the engine coolant temperature THW is higher than the first predetermined temperature. This is established when the temperature is equal to or higher than the second predetermined temperature and the downstream air-fuel ratio sensor 67 is activated. Accordingly, the sub feedback control is not established during the fuel cut and during the fuel cut return increase.

  The description will be continued assuming that the sub-feedback control condition is satisfied. Further, the power supply from the battery to the backup RAM 74 is temporarily cut off and the power supply is resumed after that because the battery of the vehicle on which the engine 10 is mounted is removed from the vehicle, and the learning value Vafsfbg is initialized. The description will be continued assuming that the value has just been returned. When the power supply from the battery to the backup RAM 74 is resumed, the CPU 71 sets the learning value Vafsfbg, the value of the learning completion flag XGK, and the value of the counter (inversion number counter) C indicating the number of inversions to “0”. It is supposed to be set. These values are all stored in the backup RAM 74.

In this case, the CPU 71 makes a “Yes” determination at step 705, sequentially performs the processing from step 710 to step 720 described below, and proceeds to step 725.
Step 710: The CPU 71 acquires an output deviation amount DVoxs that is a difference between the downstream target value Voxsref and the output value Voxs of the downstream air-fuel ratio sensor 67 according to the above equation (9). Note that the downstream target value Voxsref at this time point is set to the stoichiometric air-fuel ratio equivalent value Vst by the processing in step 810 of FIG. 8 described later.
Step 715: The CPU 71 acquires the sub feedback amount Vafsfb according to the above equation (10). Note that the integral gain Ki at this time point is set to the first integral gain KiLarge by the process of step 815 in FIG. 8 described later.
Step 720: The CPU 71 obtains a new output deviation amount integrated value SDVoxs by adding the output deviation amount DVoxs obtained in step 715 to the integrated value SDVoxs of the output deviation amount at that time.

  In step 725, the CPU 71 determines whether or not the current time is immediately after the above-described lean inversion or immediately after the above-described rich inversion. That is, the CPU 71 determines whether or not the current time is “a time immediately after the output value Voxs of the downstream air-fuel ratio sensor 67 crosses the downstream target value Voxsref set to the stoichiometric air-fuel ratio equivalent value Vst”. . If the current time point is not immediately after the above-described lean inversion or immediately after the above-described rich inversion, the CPU 71 makes a “No” determination at step 725 to directly proceed to step 795 to end the present routine tentatively.

  On the other hand, if the current time is immediately after the above-described lean inversion or immediately after the above-described rich inversion, the CPU 71 determines “Yes” in step 725 and sequentially performs the processing from step 730 to step 745 described below. Thereafter, the routine proceeds to step 795 to end the present routine tentatively.

Step 730: The CPU 71 stores the learning value Vafsfbg at that time as the pre-update learning value Vafsfbg0.
Step 735: The CPU 71 updates the learning value Vafsfbg according to the above equation (11).
Step 740: The CPU 71 calculates a change amount (update amount) ΔG of the learning value Vafsfbg according to the above equation (12).
Step 745: The CPU 71 corrects the sub feedback amount Vafsfb with the change amount ΔG according to the above equation (13).

  With the above processing, the sub feedback amount Vafsfb is updated every time a predetermined time elapses. Further, the learning value Vafsfbg of the sub feedback amount is updated every lean inversion and rich inversion, and the sub feedback amount Vafsfb is corrected.

  By the way, the CPU 71 repeatedly executes the learning completion determination routine shown in the flowchart of FIG. 8 every elapse of a predetermined time. Accordingly, when the predetermined timing is reached, the CPU 71 starts processing from step 800 and proceeds to step 805 to determine whether or not the value of the learning completion flag XGK is “0”. That is, the CPU 71 determines whether or not the learning control is incomplete (whether or not it is determined that the learning control has been completed).

  If the above assumption is followed, the learning control is incomplete, so the value of the learning completion flag XGK is “0”. Accordingly, the CPU 71 makes a “Yes” determination at step 805 to proceed to step 810 to store / set the stoichiometric air / fuel ratio equivalent value Vst in the downstream target value Voxsref. Next, the CPU 71 proceeds to step 815 to store / set the first integral gain KiLarge as the integral gain Ki.

  As a result, the sub-feedback amount Vafsfb is calculated in a state where the downstream target value Voxsref is set to the stoichiometric air-fuel ratio equivalent value Vst and the integral gain Ki is set to the first integral gain KiLarge until the learning completion determination is made. The learning value Vafsfbg is updated.

  Next, the CPU 71 proceeds to step 820, and determines whether or not the current time is immediately after the above-described lean inversion or immediately after the above-described rich inversion. If the current time is not immediately after the above-described lean inversion or immediately after the above-described rich inversion, the CPU 71 makes a “No” determination at step 820 to directly proceed to step 895 to end the present routine tentatively.

  On the other hand, if the current time is immediately after the above-described lean inversion or immediately after the above-described rich inversion, the CPU 71 determines “Yes” in step 820 and proceeds to step 825 to set the inversion number counter C by “1”. Increase. Next, the CPU 71 proceeds to step 830 and determines whether or not the value of the inversion number counter C is equal to or greater than the threshold number Cth. In this example, the learning value Vafsfbg is updated as many times as the number of inversions. Accordingly, the threshold number Cth is set to “a value where the learning value Vafsfbg is sufficiently updated and reaches an appropriate value (the learning value Vafsfbg converges) if the inversion number C reaches the threshold number Cth”. ing.

  If it is immediately after the start of learning, the value of the inversion number counter C is smaller than the threshold number Cth. In this case, the CPU 71 determines that learning has not yet been completed, and proceeds directly from step 830 to step 895 to end the present routine tentatively.

  When the predetermined time elapses, the processing from step 730 to step 745 of the routine of FIG. 7 is repeated, so that the learning value Vafsfbg approaches an appropriate value and gradually converges. Further, since the process of step 825 in FIG. 8 is repeatedly performed, the value of the inversion number counter C gradually increases and reaches the threshold number Cth.

  At this time, when the CPU 71 executes the process of step 830 in FIG. 8, the CPU 71 determines “Yes” in the step 830 (that is, determines that learning is completed), and performs steps 835 and 840 described below. After performing the processing in order, the routine proceeds to step 895, and this routine is temporarily terminated.

Step 835: The CPU 71 sets the value of the learning completion flag XGK to “1”.
Step 840: The CPU 71 corrects (increases) the sub feedback amount Vafsfb by the above-described correction amount β according to the above equation (14).

  Thus, the value of the learning completion flag XGK is set to “1”, so that the CPU 71 determines “No” in step 805 of FIG. In this case, the CPU 71 sequentially performs the processing of step 845 and step 850 described below, proceeds to step 895, and once ends this routine.

Step 845: The CPU 71 stores and sets a target rich air-fuel ratio equivalent value Vrich (a value corresponding to the target non-stoichiometric air-fuel ratio) in the downstream target value Voxsref.
Step 850: The CPU 71 stores and sets the second integral gain KiSmall as the integral gain Ki.

  As a result, after the learning completion determination is made, the sub-feedback amount Vafsfb with the downstream target value Voxsref set to the target rich air-fuel ratio equivalent value Vrich and the integral gain Ki set to the second integral gain KiSmall. Is calculated and the learning value Vafsfbg is updated (see step 710 to step 745 in FIG. 7).

If the sub-feedback control condition is not satisfied, the CPU 71 makes a “No” determination at step 705 in FIG. 7, performs the processing of step 750 and step 755 described below in order, and proceeds to step 795 to execute this routine. Is temporarily terminated.
Step 750: The CPU 71 sets the value of the sub feedback amount Vafsfb to “0”.
Step 755: The CPU 71 sets the value of the integrated value SDVoxs of the output deviation amount to “0”.

  Thereby, as is apparent from the above equation (1), the feedback control output value Vabyfc is the sum of the output value Vabyfs of the upstream air-fuel ratio sensor 66 and the learning value Vafsfbg. That is, in this case, the update of the sub feedback amount Vafsfb and the reflection to the final fuel injection amount Fi are stopped. However, at least the learning value Vafsfbg corresponding to the integral term of the sub feedback amount Vafsfb is reflected in the final fuel injection amount Fi.

As described above, the embodiment of the air-fuel ratio control apparatus for an internal combustion engine according to the present invention is
A catalyst (53) disposed in an exhaust passage of the internal combustion engine and having an oxygen storage function for storing oxygen ;
An upstream air-fuel ratio sensor (66);
A downstream air-fuel ratio sensor (67);
Sub-feedback amount calculation means (steps 710 to 720 in FIG. 7) for calculating a sub-feedback amount for matching the output value of the downstream air-fuel ratio sensor with the downstream target value that is the target value;
Learning to update the learning value of the sub-feedback amount based on the value corresponding to the steady component included in the sub-feedback amount every time the output value of the downstream air-fuel ratio sensor changes so as to cross the downstream target value. Performing (step 725 and step 735 in FIG. 7) and correcting the sub feedback amount according to the updated learning value (step 730, step 740 and step 745 in FIG. 7), learning means;
Learning completion determination means for acquiring a learning completion index value representing the degree of convergence of the learning value and determining whether the learning has been completed based on the learning completion index value (steps 820 to 835 in FIG. 8) When,
The air-fuel ratio obtained based on the output value Vabyfs of the upstream air-fuel ratio sensor, the corrected sub-feedback amount Vafsfb, and the updated learned value Vafsfbg is matched with the upstream target air-fuel ratio abyfr that is the target value. Air-fuel ratio control means for executing air-fuel ratio feedback control for controlling the air-fuel ratio of the air-fuel mixture supplied to the internal combustion engine (steps 610 to 640 in FIG. 6, steps 520 to 550 in FIG. 5),
Is an air-fuel ratio control device.

The air-fuel ratio control means includes
Based on the output value Vabyfs of the upstream air-fuel ratio sensor, the corrected sub-feedback amount Vafsfb, and the updated learned value Vafsfbg, the air-fuel ratio of the engine is made to coincide with the upstream target air-fuel ratio abyfr. Means for calculating the main feedback amount DFi and controlling the air-fuel ratio of the air-fuel mixture supplied to the engine based on the main feedback amount DFi (steps 610 to 640 in FIG. 6 and steps 520 to 550 in FIG. 5) It can also be expressed as

Furthermore, the control device
Main feedback amount calculation means for calculating a main feedback amount DFi for making the feedback control air-fuel ratio abyfsc obtained based on the output value Vabyfs of the upstream air-fuel ratio sensor coincide with the upstream target air-fuel ratio abyfr (step of FIG. 6) 610 to step 640),
Sub feedback amount calculating means (steps 710 to 720 in FIG. 7) for calculating a sub feedback amount for making the output value of the downstream air-fuel ratio sensor coincide with the downstream target value;
The learning means (steps 725 to 735 in FIG. 7, steps 740 and 745 in FIG. 7);
The learning completion determining means (steps 820 to 835 in FIG. 8);
Air-fuel ratio control means for controlling the air-fuel ratio of the air-fuel mixture supplied to the engine based on the calculated main feedback amount, the corrected sub-feedback amount, and the updated learned value (step 610 in FIG. 6). Steps 520 to 550 in FIG.
The downstream target value is set to a value Vst corresponding to a theoretical air-fuel ratio in a period before the learning completion determination unit determines that the learning is completed, and the learning is completed by the learning completion determination unit. A downstream target value setting that sets the downstream target value to a value corresponding to a target non-stoichiometric air-fuel ratio (for example, target rich air-fuel ratio Vrich) that is an air-fuel ratio different from the stoichiometric air-fuel ratio in the period after the determination. Means (step 805, step 810 and step 845 in FIG. 8);
It can also be said that the air-fuel ratio control device is provided with

  According to the control device configured as described above, the downstream target value Voxsref is set to the value Vst corresponding to the theoretical air-fuel ratio before the learning value Vafsfbg is determined to be completed. Therefore, since the temporal average value of the air-fuel ratio of the engine is substantially the stoichiometric air-fuel ratio, the air-fuel ratio of the gas flowing out from the upstream side catalyst 53 oscillates around the stoichiometric air-fuel ratio. As a result, “the frequency with which lean inversion and rich inversion occur” increases more than when the downstream target value is set to a value Vrich corresponding to the target rich air-fuel ratio (a value corresponding to the target non-stoichiometric air-fuel ratio). That is, the learning value update opportunity increases as the number of inversions C increases early, so that learning can be completed early.

Further, the learning completion determining means includes
The number of inversions C, which is the number of times the output value of the downstream air-fuel ratio sensor changes so as to cross the value corresponding to the theoretical air-fuel ratio, is acquired as the learning completion index value, and the acquired number of inversions C is a predetermined threshold value. When it is determined that the number of times Cth or more has been reached, it is determined that the learning has been completed (steps 820 to 835 in FIG. 8).

  The inversion number C increases according to the number of updates of the learning value Vafsfbg. Therefore, according to the control device, the learning completion index value indicating the degree of convergence of the learning value can be obtained accurately and easily.

Further, the sub-feedback amount calculating means includes
When the learning completion determination means determines that the learning is completed, the learning value is corrected by an amount β corresponding to the difference between the theoretical air-fuel ratio and the target non-stoichiometric air-fuel ratio (target rich air-fuel ratio in this example). (Steps 830 and 840 in FIG. 8).

  Therefore, the learning value obtained before the learning completion determination in which the downstream target value Voxsref is the theoretical air-fuel ratio equivalent value Vst is used as the target rich air-fuel ratio equivalent value Vrich where the downstream target value Voxsref is the target non-stoichiometric air-fuel ratio equivalent value. After the learning completion determination, it can be converted into a learning value having an appropriate value.

  Further, the control device sets a first condition as a sub-feedback control condition before it is determined that the learning is completed, and after the determination that the learning is completed, a condition ( In this example, there is provided sub-feedback control condition setting means for setting, as the sub-feedback control condition, a second condition to which a fuel cut control condition is not satisfied.

  In other words, in the above control device, the fuel cut control condition before the learning completion determination is actually a condition obtained by adding the condition that the learning is completed to the fuel cut control condition after the learning completion determination (FIG. 5). (See Step 510 and Step 560.) The sub-feedback control condition includes a condition that fuel cut is not being performed. Accordingly, the sub feedback control condition after the learning completion determination is obtained by adding a condition that the fuel cut control condition is not actually established to the sub feedback control condition before the learning completion determination.

The sub feedback amount calculating means is configured to calculate the sub feedback amount only when the sub feedback control condition is satisfied (see step 705 in FIG. 7).
The learning means is configured to perform “correction according to the learning value of the learning and the sub feedback amount” only when the sub feedback control condition is satisfied (step 705 in FIG. 7 is performed). reference.).

  According to this, the sub-feedback control is executed because the sub-feedback control condition (first condition) that is more relaxed than the sub-feedback control condition (second condition) after completion of learning is effectively set before learning is completed. It becomes easy to be done. As a result, learning opportunities increase, and the learning completion time can be advanced.

Furthermore, the air-fuel ratio control apparatus for the internal combustion engine is
The fuel cut control for stopping the fuel supply to the engine is not executed even if a predetermined fuel cut condition is satisfied in a period before the learning completion determining means determines that the learning is completed. The fuel cut control means for executing the fuel cut control when the fuel cut condition is satisfied in a period after the learning completion determination means determines that the learning is completed (step 510 and FIG. 5). (See step 560.)

  According to this, the fuel cut control is not executed in the period before the learning is completed, and further, the increase control after returning from the fuel cut is not executed. Accordingly, learning opportunities can be increased. Moreover, since fuel cut control is performed after completion of learning, it is possible to protect the catalyst and improve emissions.

  The present invention is not limited to the above embodiments, and various modifications can be adopted within the scope of the present invention. For example, the control device performs basic fuel injection based on the in-cylinder fuel supply amount deviation DFc, which is a value obtained by subtracting the in-cylinder fuel supply amount Fc (k−N) from the target in-cylinder fuel supply amount Fcr (k−N). The main feedback amount DFi for correcting the amount Fbase has been calculated. On the other hand, the main feedback amount for correcting the basic fuel injection amount Fbase may be calculated from the air-fuel ratio difference between the feedback control air-fuel ratio abyfsc and the upstream target air-fuel ratio abyfr.

  Further, the sub-feedback control of the control device apparently corrects the air-fuel ratio abyfs detected by the upstream air-fuel ratio sensor 66 so that the output value Voxs of the downstream air-fuel ratio sensor 67 matches the downstream target value Voxsref. (Refer to the above formula (1).) On the other hand, in the sub feedback control, as disclosed in JP-A-6-010738, the air-fuel ratio correction coefficient created based on the output value of the upstream air-fuel ratio sensor 66 is changed to the downstream air-fuel ratio sensor 67. The output value Voxs may be changed based on the sub-feedback amount obtained by proportional integration.

In addition, the air-fuel ratio control apparatus according to the present invention includes an upstream air-fuel ratio sensor 66 as disclosed in JP 2007-77869 A, JP 2007-146661 A, JP 2007-162565 A, and the like. The difference between the upstream air-fuel ratio abyfs obtained based on the output value Vabyfs and the upstream target air-fuel ratio abyfr is high-pass filtered to calculate the main feedback amount KFmain, and the downstream air-fuel ratio sensor 67 output value Voxs and downstream The sub feedback amount Fisub may be obtained by performing proportional integration processing on a value obtained by performing low-pass filter processing on the deviation from the side target value Voxsref. In this case, as shown in the following equation (15), these feedback amounts are used to correct the basic fuel injection amount Fbase in a form independent of each other, thereby obtaining the final fuel injection amount Fi. May be.
Fi = KFmain · Fbase + Fisub (15)

  Furthermore, the routine shown in FIG. 8 may be configured to be executed only when the sub feedback control condition is satisfied (during the sub feedback control in which the sub feedback amount Vafsfb is updated). Further, in the routine of FIG. 7, a step of determining whether “the value of the learning completion flag XGK is“ 0 ”” is inserted between step 720 and step 725, and the value of the learning completion flag XGK is “ It is also possible to configure so that the learning value Vafsfbg is updated by proceeding to Step 730 to Step 745 only when it is “0”. That is, after the time when the value of the learning completion flag XGK is set to “1”, the update of the learning value Vafsfbg may be stopped until the value of the learning completion flag XGK is next set to “0”.

  In addition, the control device according to the above embodiment sets the downstream target value Voxsref to a value Vrich corresponding to the target rich air-fuel ratio in a period after it is determined that learning is completed. Instead, the present invention aims to reduce the downstream target value in a period after the learning is determined to be complete for the purpose of reducing the exhaust gas purification characteristics of the catalyst and the exhaust odor of the gas discharged from the engine. The present invention can also be applied to a device that sets Voxsref to a value corresponding to a target lean air-fuel ratio that is leaner than the stoichiometric air-fuel ratio. That is, the present invention is applied to an apparatus that sets the downstream target value Voxsref to a value corresponding to a target non-stoichiometric air-fuel ratio that is an air-fuel ratio other than the stoichiometric air-fuel ratio in a period after it is determined that learning has been completed. .

  Further, the control device according to the present invention, for example, continues the change of the learning value Vafsbfbg and the correction of the sub-feedback amount Vafsfb (that is, learning) in steps 725 to 745 in FIG. It is also possible to configure to be performed after the established time point. In this case, the learning opportunity (the learning value Vafsbfbg is changed by setting the predetermined delay time Ta to a time shorter than after the learning is determined to be completed before the learning is determined to be completed. It is desirable to increase opportunities.

1 is a schematic view of an internal combustion engine to which an air-fuel ratio control apparatus according to an embodiment of the present invention is applied. 2 is a graph showing a relationship between an air-fuel ratio and an output value of an upstream air-fuel ratio sensor shown in FIG. 2 is a graph showing a relationship between an air-fuel ratio and an output value of a downstream air-fuel ratio sensor shown in FIG. It is a time chart which shows the action | operation of the control apparatus shown in FIG. 2 is a flowchart showing a fuel injection control routine executed by a CPU shown in FIG. 2 is a flowchart showing a routine executed by the CPU shown in FIG. 1 to calculate a main feedback amount. 2 is a flowchart showing a routine executed by the CPU shown in FIG. 1 to calculate a sub feedback amount and a learning value. It is the flowchart which showed the routine which CPU shown in FIG. 1 performs in order to perform learning completion determination.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Internal combustion engine, 20 ... Cylinder block part, 21 ... Cylinder, 22 ... Piston, 25 ... Combustion chamber, 30 ... Cylinder head part, 31 ... Intake port, 32 ... Intake valve, 34 ... Exhaust port, 35 ... Exhaust valve, 37 ... Spark plug, 39 ... Injector, 40 ... Intake system, 41 ... Intake pipe, 43 ... Throttle valve, 50 ... Exhaust system, 51 ... Exhaust manifold, 52 ... Exhaust pipe, 53 ... Upstream catalyst, 54 ... Downstream catalyst , 61 ... hot-wire air flow meter, 66 ... upstream air-fuel ratio sensor, 67 ... downstream air-fuel ratio sensor, 70 ... electric control device, 71 ... CPU, 74 ... backup RAM.

Claims (3)

A catalyst provided in the exhaust passage of the internal combustion engine and having an oxygen storage function for storing oxygen ;
An upstream air-fuel ratio sensor that is disposed in a portion upstream of the catalyst in the exhaust passage and outputs an output value corresponding to the air-fuel ratio of the gas flowing through the disposed portion;
A downstream air-fuel ratio sensor that is disposed in a portion downstream of the catalyst in the exhaust passage and outputs an output value corresponding to the air-fuel ratio of the gas flowing through the disposed portion;
Sub-feedback amount calculating means for calculating a sub-feedback amount for matching the output value of the downstream air-fuel ratio sensor with the downstream target value that is the target value;
Learning to update the learning value of the sub-feedback amount based on the value corresponding to the steady component included in the sub-feedback amount every time the output value of the downstream air-fuel ratio sensor changes so as to cross the downstream target value. Learning means for performing and correcting the sub feedback amount according to the updated learning value,
Learning completion determination means for acquiring a learning completion index value representing a degree of convergence of the learning value and determining whether the learning is completed based on the learning completion index value;
The internal combustion engine so that the air-fuel ratio obtained based on the output value of the upstream air-fuel ratio sensor, the corrected sub-feedback amount, and the updated learned value matches the upstream target air-fuel ratio that is the target value. Air-fuel ratio control means for performing air-fuel ratio feedback control for controlling the air-fuel ratio of the air-fuel mixture supplied to
An air-fuel ratio control apparatus for an internal combustion engine comprising:
The downstream target value is set to a value corresponding to the theoretical air-fuel ratio in a period before the learning completion determination unit determines that the learning is completed, and the learning completion determination unit determines that the learning is completed. An air-fuel ratio control device comprising downstream target value setting means for setting the downstream target value to a value corresponding to a target non-stoichiometric air-fuel ratio that is an air-fuel ratio different from the stoichiometric air-fuel ratio in a period after the operation.
The air-fuel ratio control apparatus for an internal combustion engine according to claim 1,
The learning completion judging means
The number of inversions, which is the number of times the output value of the downstream air-fuel ratio sensor changes so as to cross a value corresponding to the theoretical air-fuel ratio, is acquired as the learning completion index value, and the acquired number of inversions is equal to or greater than a predetermined threshold number of times. An air-fuel ratio control apparatus configured to determine that the learning has been completed when it is determined that the learning is completed. The air-fuel ratio control apparatus for an internal combustion engine according to claim 1 or 2,
The sub feedback amount calculating means includes
An air-fuel ratio control apparatus configured to correct the learning value by an amount corresponding to a difference between the theoretical air-fuel ratio and the target non-stoichiometric air-fuel ratio when the learning completion determining means determines that the learning is completed.

Images