JP3217682B2 - Fuel injection control device for internal combustion engine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION relates to a fuel injection control apparatus of the invention is an internal combustion engine, more specifically corrects the target air-fuel ratio in the catalyst window, O ₂ storage effect better catalytic purification to improve the catalytic device To achieve the rate.

[0002]

2. Description of the Related Art In a fuel injection control device for an internal combustion engine, a purification rate of a catalyst device provided in an exhaust system is maximized near a stoichiometric air-fuel ratio. It is known to perform feedback control of a fuel injection amount so as to obtain an air-fuel ratio.

Further, recently, as disclosed in Japanese Patent Application Laid-Open No. 3-185244, a first oxygen concentration sensor (wide-range air-fuel ratio sensor) is disposed upstream of a catalyst and a second oxygen concentration sensor is disposed downstream thereof. A sensor (O ₂ sensor) is arranged, a target air-fuel ratio is set so as to obtain an optimum purification rate in the catalyst window according to the output of the second sensor, and the target air-fuel ratio is set in accordance with the target air-fuel ratio and the output of the first sensor. There is also proposed a technique for controlling the fuel injection amount by using the same. In this conventional technique, a control target is modeled, and an optimal regulator is designed to control a fuel injection amount.

[0004]

However, in the above-described prior art, the change in the target air-fuel ratio is made to follow the target value by feedback control.
Since it is impossible to follow a change in dynamic characteristics due to a change over time or a variation in solids of the internal combustion engine, there has been a disadvantage that optimal control performance cannot be obtained. This is because the behavior of the air-fuel ratio is not adaptively compensated in the above-described related art.

Accordingly, an object of the present invention is to solve the above-mentioned drawbacks of the prior art, and to adaptively compensate for the behavior of the air-fuel ratio, so that the target value determined based on the output of the second air-fuel ratio detecting means. It is an object of the present invention to provide a fuel injection control device for an internal combustion engine that controls fuel injection so that the air-fuel ratio instantaneously matches.

[0006]

According to one aspect of the present invention, there is provided a fuel injection amount control means for controlling a fuel injection amount of an internal combustion engine, and an operation state of the internal combustion engine.
Fuel injection amount calculation means for calculating the fuel injection amount according to the state,
A first air-fuel ratio detecting unit disposed upstream of a catalyst device disposed in an exhaust system of the internal combustion engine, the first air-fuel ratio detecting unit detecting an air-fuel ratio of exhaust gas discharged from the internal combustion engine; as the air-fuel ratio to be detected equal to the target air-fuel ratio,
The fuel injection correction amount calculated by the fuel injection amount calculation means is supplemented.
Fuel injection correction to calculate the fuel injection correction coefficient to correct
The fuel injection control device for an internal combustion engine, comprising: a coefficient calculating unit; and a second air-fuel ratio detecting unit disposed downstream of the catalyst device and detecting an air-fuel ratio of exhaust gas passing through the catalyst device. Injection correction coefficient calculation means,
In accordance with the air-fuel ratio detected by the second air-fuel ratio detecting means,
And the target air-fuel ratio correction means for correcting the serial target air-fuel ratio, so that the air-fuel ratio detected by the first air-fuel ratio detection means coincides with said corrected <br/> target air-fuel ratio, the fuel injection correction coefficient The adaptive controller to be calculated and the adaptive parameter to be input to the adaptive controller are detected by the first air-fuel ratio detecting means.
An adaptive parameter adjusting mechanism for adjusting based on the detected air-fuel ratio and the past value of the adaptive parameter is provided.

According to a second aspect of the present invention, the catalyst device has a multi-stage catalyst bed, and the second air-fuel ratio detecting means is arranged between the multi-stage catalyst beds.

According to a third aspect of the present invention, a filter is connected to the first air-fuel ratio detecting means.

According to a fourth aspect of the present invention, a filter means is connected to the second air-fuel ratio detecting means.

According to a fifth aspect of the present invention, the filter means is a low-pass filter.

[0011]

[Action] In the fuel injection control apparatus for an internal combustion engine according to claim 1, wherein can be adaptively compensate the dynamic behavior of the air-fuel ratio due to change over time or a solid dispersion of the inner combustion engine, the second The air-fuel ratio can be instantaneously matched with a target value determined based on the air-fuel ratio detected by the air-fuel ratio detecting means.

Here, the "adaptive controller" is a controller that takes into account the dynamic behavior of the controlled object (internal combustion engine), and in the embodiment, compensates for the dynamic behavior of the controlled object. To do so, it consists of a controller described in recurrence form. More specifically, since it is of the STR type, it can be defined as an adaptive controller having a recurrence type adaptive parameter adjustment mechanism at the input of the controller.

According to a second aspect of the present invention, the catalyst device has a multi-stage catalyst bed, and the second air-fuel ratio detecting means is arranged between the multi-stage catalyst beds. Compared to the case where it is located downstream of the catalyst device,
The time during which the output is inverted is shortened, and the detection accuracy and, consequently, control accuracy are improved. Further, with this configuration, even if the capacity of the catalyst device is increased, the detection accuracy,
As a result, the control accuracy does not decrease.

According to a third aspect of the present invention, a filter is connected to the first air-fuel ratio detecting means.
By appropriately selecting the frequency characteristics of the filter, noise can be removed, detection accuracy increases, and controllability improves.

According to a fourth aspect of the present invention, the filter means is connected to the second air-fuel ratio detecting means.
By appropriately selecting the frequency characteristics of the filter, the response time can be optimized, the detection accuracy increases, and the controllability improves.

According to a fifth aspect of the present invention, the filter means is configured to be a low-pass filter.
By optimizing the frequency characteristics of the filter, noise can be reliably removed, or the response time can be optimized, so that detection accuracy is improved and controllability is improved.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a fuel injection control device for an internal combustion engine according to the present application will be described below with reference to the accompanying drawings.

FIG. 1 is an overall view schematically showing the apparatus.

In FIG. 1, reference numeral 10 denotes an OHC in-line four-cylinder internal combustion engine, and intake air introduced from an air cleaner 14 disposed at the tip of an intake pipe 12 is subjected to surge while its flow rate is adjusted by a throttle valve 16. The gas flows into the first to fourth cylinders via two intake valves (not shown) via the tank 18 and the intake manifold 20. An injector 22 is provided near an intake valve (not shown) of each cylinder to inject fuel. The air-fuel mixture injected and integrated with the intake air is ignited by an ignition plug (not shown) in each cylinder and burns to drive a piston (not shown).

The exhaust gas after combustion is discharged to an exhaust manifold 24 via two exhaust valves (not shown), and a first catalyst device (three-way catalyst) 28 and a second catalyst 28 via an exhaust pipe 26. The catalyst is purified by the catalyst device (three-way catalyst) 30 and discharged outside the engine. As described above, the throttle valve 16 is mechanically separated from the accelerator pedal (not shown), and is controlled via the pulse motor M to an opening degree corresponding to the depression amount of the accelerator pedal and the operating state. The intake pipe 12 is provided with a bypass passage 32 near the position of the throttle valve 16 for bypassing the throttle valve 16.

Here, the internal combustion engine 10 is provided with an exhaust gas recirculation mechanism 100 for recirculating exhaust gas to the intake side.

Referring to FIG. 2, one end 121a of the exhaust gas recirculation path 121 of the exhaust gas recirculation
The other end 121b communicates with the downstream side of the throttle valve 16 (not shown in FIG. 2) of the intake pipe 12 on the upstream side of the first catalyst device 28 (not shown in FIG. 2). This exhaust gas recirculation path 1
In the middle of 21, an exhaust gas recirculation valve (recirculation gas control valve) 122 for adjusting the amount of exhaust gas recirculation and a volume chamber 121c are provided. The exhaust gas recirculation valve 122 is an electromagnetic valve having a solenoid 122a. The solenoid 122a is connected to a control unit (ECU) 34, which will be described later.
The valve opening is changed linearly by the output from the controller.
The exhaust gas recirculation valve 122 is provided with a lift sensor 123 for detecting the opening degree of the valve, and the output is sent to the control unit 34.

Further, the connection between the intake system of the internal combustion engine 10 and the fuel tank 36 is established, and the canister purge mechanism 200 is connected.
Is provided.

As shown in FIG. 3, the canister-purge mechanism 200 includes the upper part of the sealed fuel tank 36 and the intake pipe 12.
Of the steam supply passage 221 and the adsorbent 231, which is formed between the downstream side of the throttle valve 16 and the canister 2.
23, and a purge passage 224. Steam supply passage 2
21, a two-way valve 222 is mounted, a purge control valve 225 is provided in the middle of the purge passage 224, a flow meter 226 for detecting a flow rate of a mixture containing fuel vapor flowing through the purge passage 224, Is provided with an HC concentration sensor 227 for detecting the HC concentration of the liquid crystal. The purge control valve (electromagnetic valve) 225 is connected to the control unit 34 as described later, and is controlled according to a signal from the control unit 34 to linearly change the valve opening amount.

According to the canister purge mechanism, the fuel vapor (fuel vapor) generated in the fuel tank 36 is
When a predetermined set amount is reached, the positive pressure valve of the two-way valve 222 is pushed open, flows into the canister 223, and adsorbent 2
Adsorbed by 31 and stored. When the purge control valve 225 is opened by an opening amount corresponding to the duty ratio of the on / off control signal from the control unit 34, the canister 22
The evaporative fuel temporarily stored in 3 is sucked into the intake pipe 12 through the purge control valve 225 together with the outside air sucked from the outside air intake port 232 by the negative pressure in the suction pipe 12 and sent to each cylinder. When the fuel tank 36 is cooled by the outside air and the negative pressure in the fuel tank increases, the two-way valve 22
The second negative pressure valve is opened, and the evaporated fuel temporarily stored in the canister 223 is returned to the fuel tank 36.

Further, the internal combustion engine 10 includes a so-called variable valve timing mechanism 300 (shown as V / T in FIG. 1). The variable valve timing mechanism 300 is described in, for example, Japanese Patent Application Laid-Open No. 2-275043, and has two types of valve timing V / T of the engine shown in FIG. 4 in accordance with operating conditions such as the engine speed Ne and the intake pressure Pb. Switch between LoV / T and HiV / T. However, the mechanism itself is a well-known mechanism, and thus the description is omitted. Note that the switching of the valve timing characteristics includes an operation of stopping one of the two intake valves.

As shown in FIG. 1, a crank angle sensor 40 for detecting a crank angle position of a piston (not shown) is provided in a distributor (not shown) of the internal combustion engine 10, and an opening degree of the throttle valve 16 is provided. And an absolute pressure sensor 44 for detecting the intake pressure Pb downstream of the throttle valve 16 as an absolute pressure. At an appropriate position of the internal combustion engine 10, the atmospheric pressure Pa
, An intake air temperature sensor 48 for detecting the temperature of the intake air is provided upstream of the throttle valve 16, and a water temperature sensor 50 for detecting the engine cooling water temperature at an appropriate position of the engine. Is provided. Also,
Valve timing (V / T) for detecting the selected valve timing characteristics of the variable valve timing mechanism 300 via hydraulic pressure
) A sensor 52 (not shown in FIG. 1) is also provided.

Further, in the exhaust system, the exhaust manifold 2
A wide-range air-fuel ratio sensor 54 is provided as a first air-fuel ratio detecting means in the exhaust system collecting portion on the downstream side of the first catalyst device 28 on the downstream side of the first catalyst device 28, and the second air-fuel ratio is provided on the downstream side thereof. An O ₂ sensor 56 is provided as detection means. here,
The capacity of the first catalyst device 28 is about 1 liter, and the capacity of the second catalyst device 30 is about 1.7 liter. The capacities of the catalyst devices 28 and 30 are set to optimal capacities in consideration of the purifying performance and the temperature rising characteristics of the catalyst devices.

As shown in FIG. 5, the first catalyst device 28 has a multi-stage, two-stage catalyst bed (CAT) in the illustrated example.
Floor) (carrier), and the O ₂ sensor 56 includes first and second
May be arranged between the CAT floors. In that case, the capacity of the first CAT bed is about 1 liter,
The capacity of the AT floor is also about 1 liter. As a result, FIG.
The first of the entire catalyst unit 28 has a capacity of about 2 liters of O ₂ sensor that is provided at the position of the, the O ₂ sensor is substantially downstream of the catalytic converter of approximately 1 liter shown in The time required for the output to be reversed is shorter than that provided when the catalyst device is provided downstream of a 2-liter catalyst device. Therefore, the O ₂ sensor 5
Based on the output of No. ₆ , the control accuracy in performing the minute control of the air-fuel ratio in the catalyst window (this is called "MIDO2 control" in this specification) is improved as described later.

A filter 58 is connected to the next stage of the wide area air-fuel ratio sensor 54. The second filter 60 is also connected to the next stage of the O ₂ sensor 56. The sensor output and the filter output are sent to the control unit 34.

FIG. 6 is a block diagram showing details of the control unit 34. The output of the wide-range air-fuel ratio sensor 54 is input to a first detection circuit 62, where appropriate linearization processing is performed to detect linear characteristics proportional to the oxygen concentration in the exhaust gas in a wide range from lean to rich. (Hereinafter referred to as “LAF”).
Sensor)). The output of the O ₂ sensor 56 is the second
As shown in FIG. 7, the air-fuel ratio of the air-fuel mixture supplied to the internal combustion engine 10 is changed to the stoichiometric air-fuel ratio (λ =
For 1), a detection signal indicating rich or lean is output.

The output of the first detection circuit 62 is input to the CPU via the multiplexer 66 and the A / D conversion circuit 68. CPU is CPU core 70, ROM 72, R
More specifically, the output of the first detection circuit 62 is A / A at every predetermined crank angle (for example, 15 degrees).
The data is D-converted and sequentially stored in one of the buffers in the RAM 74. As shown in FIG. 47, Nos. 0 to 11 are stored in the twelve buffers later. Is appended. Similarly, the output of the second detection circuit 64 and the output of the analog sensor such as the throttle opening sensor 42 are also taken into the CPU via the multiplexer 66 and the A / D conversion circuit 68 and stored in the RAM 74.

The output of the crank angle sensor 40 is shaped by a waveform shaping circuit 76, the output value is counted by a counter 78, and the count value is input to the CPU.
In the CPU, the CPU core 70 calculates a control value according to a command stored in the ROM 72 as described later, and drives the injector 22 of each cylinder via the drive circuit 82. Further, the CPU core 70 is connected to the solenoid valve 90 (the bypass passage 32 for adjusting the secondary air amount) through the drive circuits 84, 86, 88.
Opening and closing), and the above-described exhaust recirculation control solenoid valve 122.
And drives the canister purge control solenoid valve 225. In FIG. 6, illustration of the lift sensor 123, the flow meter 226, and the HC concentration sensor 227 is omitted.

FIG. 8 is a functional block diagram for explaining the operation of the fuel injection control device according to the embodiment.

As shown in the figure, the fuel injection control device according to the embodiment includes an observer (shown as OBSV in the figure) for estimating the air-fuel ratio of each cylinder from the output of a single LAF sensor 54 and a LAF sensor. An adaptive controller (Self Tuning Regu) that inputs the output of
lator type adaptive controller. STR in the figure).

The output VO ₂ M of the O ₂ sensor 56 is supplied to a target air-fuel ratio correction block (KCM in FIG.
Is input to indicate a D correction), O ₂ target value of the sensor (Vre
fM), the target air-fuel ratio correction coefficient KCMDM is determined. On the other hand, as described later, the basic fuel injection amount TiM-F is calculated based on the change in the effective opening area of the throttle valve, and the target air-fuel ratio correction coefficient KCMDM is calculated based on various corrections including an EGR or a canister purge correction coefficient described later. Coefficient KTOTAL
At the same time, the basic fuel injection amount TiM-F is multiplied (the multiplication symbol is used in place of the addition point in the figure to indicate the multiplication symbol) and corrected to obtain the required fuel injection amount Tcyl.

The corrected target air-fuel ratio KCMD is input to an adaptive controller STR and a PID controller (denoted by PID in the figure), and a feedback correction coefficient KSTR or KLAF is provided in accordance with the difference from the LAF sensor output as described later. Is required,
Either is multiplied by the required fuel injection amount Tcyl according to the operating state via a changeover switch (shown as a changeover SW in the figure),
The output fuel injection amount Tout is determined. Output fuel injection amount T
Out is subjected to adhesion correction as described later, and is supplied to the internal combustion engine 10.

That is, the air-fuel ratio is controlled to the target air-fuel ratio based on the output of the LAF sensor 54, and the MI value near the target value, that is, near the so-called catalyst window.
That is, DO ₂ control is performed. To further explain this, there is an O ₂ storage effect in which O ₂ is stored during the passage of a slightly lean exhaust gas as a function of the catalyst device. However, if O ₂ is saturated in the catalyst device, the purification rate is reduced. It is necessary to supply a slightly rich exhaust gas to release O ₂ . When the release of O ₂ is completed, a slightly lean exhaust gas is sent again, and this operation is repeated, whereby the purification rate of the catalyst device can be maximized.
MIDO ₂ control intends this.

In order to further improve the purification rate in the MIDO ₂ control, the air-fuel ratio in front of the catalyst device is adjusted to the target air-fuel ratio in as short a time as possible from the inversion of the output of the O ₂ sensor 56 after the catalyst device. It is necessary for the detected air-fuel ratio KACT to be equal to the target air-fuel ratio KCMD, but the target air-fuel ratio correction coefficient
Simply multiplying by KCMDM results in a delay in the response of the engine, so that the target air-fuel ratio KCMD becomes the smoothed detected air-fuel ratio KACT.

In order to improve this, the response of the detected air-fuel ratio KACT is dynamically compensated from the target air-fuel ratio KCMD, and more specifically, the correction coefficient KSTR (the output of the adaptive controller STR output) dynamically compensates the target air-fuel ratio KCMD. ) Is multiplied. By doing so, the detected air-fuel ratio KACT quickly converges to the target air-fuel ratio KCMD, and the catalyst purification rate can be improved. In this specification, both the target value KCMD and the actual value (detected value) KACT of the air-fuel ratio are actually shown as equivalent ratios, that is, Mst / M = 1 / λ (M
st: theoretical air-fuel ratio, M = A / F (A: air consumption, F: fuel consumption), λ: excess air ratio).

Here, the description of the filter will be supplemented.

The illustrated device has a multiple feedback configuration in which a plurality of control methods are provided in parallel using a single sensor output. More specifically, since the configuration is such that multiple feedback and a plurality of control methods are switched, the cutoff frequency characteristics of the filter are set according to the control method.

Specifically, the output of the LAF sensor 54 is
It takes about 400 ms for a 100% response. However, as it is, there is much noise of high frequency components, and controllability deteriorates. Therefore, when passing through a low-pass filter of 500 Hz, harmful high-frequency component noise can be removed,
It was found that the response characteristics were hardly deteriorated.
Then, when the filter frequency was lowered to 4 Hz, the high-frequency noise was further greatly reduced. Also, the time required for 100% response was stable. However, the response characteristic in this case is slightly slower than when the signal does not pass through a filter or passes through a low-pass filter of 500 Hz.
It took about 400 ms or more for 0% response.

From the above, in the case of the embodiment, the filter 5
8 is a low-pass filter having a cut-off frequency characteristic of 500 Hz, and 500 Hz is input to the observer.
The output of the low-pass filter 58 is used as it is. This is because the observer itself does not control the detected air-fuel ratio KACT to converge to the target air-fuel ratio KCMD, and the PID controller calculates the air-fuel ratio variation between cylinders based on the air-fuel ratio of each cylinder estimated by the observer. Because of the absorption configuration, even if the response time of the sensor is not very stable, it does not significantly affect the estimation result, and rather the quicker the response time, the better the controllability. is there.

On the other hand, the filter 92 (shown only in FIG. 8) connected before the input to the adaptive controller STR is a low-pass filter having a cut-off frequency characteristic of 4 Hz. That is, a device that performs dead beat control such as STR operates so as to faithfully compensate for a delay with respect to the detected air-fuel ratio. Therefore, when the noise or response time of the detected air-fuel ratio changes, the control performance itself is changed. Affect. Therefore, the filter 92 has 4H
A low-pass filter having a cut-off frequency characteristic of z.
The filter 93 connected before the input of the PID controller attaches importance to the response time, and has a filter 9 in cutoff frequency characteristics.
2 or more, 200 Hz in the case of the embodiment
And The filter 6 connected to the O ₂ sensor 56
In the case of 0, the response time is inherently much higher than that of the LAF sensor due to the characteristics of the O ₂ sensor.
A low-pass filter having a cutoff frequency characteristic of about 0 Hz is used.

Hereinafter, the operation of the apparatus according to the present application will be described with reference to the block diagram of FIG.

First, the basic fuel injection amount TiM-F is calculated.

As described above, the basic (required) fuel injection amount can be optimally determined over all operating states including the transient operating state, based on the change in the effective opening area of the throttle valve.

FIG. 9 is a flow chart showing the operation of calculating the basic fuel injection amount TiM-F, and FIG. 10 is a block diagram for explaining the calculation of the flow chart of FIG. 9, which will be described with reference to FIG. Prior to this, a method of estimating the air passing through the throttle and the air flowing into the cylinder by a method of approximating the model using the concept of the fluid dynamics model premised on this method will be described. The details are described in Japanese Patent Application No. 6-197,238 previously proposed by the present applicant, and will be briefly described below.

That is, as shown in FIG. 11, a projection area S of the throttle (a projection area of the throttle in the longitudinal direction of the intake pipe) S is obtained from the throttle opening θTH in accordance with preset characteristics. On the other hand, as shown in FIG.
The coefficient C (the product of the flow rate coefficient α and the gas expansion correction coefficient ε) is determined according to another characteristic set in advance with respect to the intake pressure Pb, and the two are multiplied to determine the effective opening area A of the throttle. Since the throttle does not become a throttle in the so-called full throttle region, the throttle fully open region is determined as a critical value for each engine speed, and when the detected throttle opening exceeds that, the critical value is determined as the throttle opening. I do. In addition, the air pressure is corrected for this, but the description is omitted.

Next, the amount of air Gb in the chamber is determined from the equation shown in Equation 1 based on the equation of state of gas, and the amount of air ΔGb that has been charged into the chamber this time is determined from the change in chamber pressure ΔP according to Equation 2. If it is assumed that the amount of air charged into the chamber this time is not naturally taken into the cylinder combustion chamber, the amount Gc of cylinder intake air per unit time ΔT is given by
Can be represented as shown in the following equation. Here, the “chamber” is not only a part equivalent to a so-called surge tank,
It means all parts between the downstream of the throttle and the intake port. “Chamber” means an effective volume that actually serves as a chamber. In this specification, k indicates a sampling time.

[0052]

(Equation 1)

[0053]

(Equation 2)

[0054]

(Equation 3)

On the other hand, as shown in FIG. 13, the above-mentioned ROM 72 stores the fuel injection amount Tim in the steady operation state.
The ap is set in advance so as to be searched from the engine speed Ne and the intake pressure Pb based on the so-called speed density method, and is stored in a map. The fuel injection amount Timap is set according to the target air-fuel ratio KCMD determined according to the engine speed Ne and the intake pressure Pb.
As shown in FIG. 4, the target air-fuel ratio KCMD, more specifically, the basic value KBS is also determined by the engine speed Ne and the intake pressure Pb.
Is stored in advance as a map so that it can be freely searched. However, the correction of the fuel injection amount Timap based on the target air-fuel ratio is MID
No modification is made here as it relates to O ₂ control. M
It will be described later modified by the target air-fuel ratio, including the IDO ₂ control. Note that the fuel injection amount Timap is directly set in units of the valve opening time of the injector 22.

Here, focusing on the relationship between the fuel injection amount Timap obtained by searching the map and the above-mentioned throttle passing air amount Gth, under a certain condition (the engine speed Ne1 and the intake pressure Pb1) in the steady operation state. The fuel injection amount Timap1 determined by the map search is as shown in Expression 4.

[0057]

(Equation 4)

Here, the amount of air passing through the throttle during the transient operation state can be expressed from the amount of air passing through the throttle in the steady state according to the change in the effective opening area of the throttle.
Specifically, it can be expressed by using the ratio of the effective opening area of the throttle valve in the steady state to the effective opening area of the throttle valve in the transient state. This is described in detail in the aforementioned Japanese Patent Application No. Hei 6-197,238.

That is, assuming that the current effective opening area of the throttle valve is A and the effective opening area of the throttle valve in the steady operation state is A1, the effective opening area A1 of the throttle valve in the steady operation state is It was estimated that it could be grasped as a first-order delay of the effective opening area A, and it was verified through simulation that it could be confirmed as shown in FIG. That is, if the first-order delay of A is called "ADELAY", it can be seen that A1 and ADELAY have almost the same value. Therefore, if the model is approximated using the concept of the fluid dynamics model, A / "the first-order lag" may be used. As shown in FIG. 16, in the transient operation state, the moment the throttle is opened, the pressure difference before and after the throttle is large, so the amount of air passing through the throttle flows all at once, and gradually falls to a steady state. It was thought that Gth could be represented by this ratio A / ADELAY. As shown in the lower part of FIG. 17, this ratio becomes 1 in the steady operation state. Hereinafter, this ratio is referred to as “RATIO-A”.

Further, focusing on the relationship between the effective opening area of the throttle and the throttle opening θTH, the effective opening area greatly depends on the throttle opening, and as shown in FIG. It should change almost following the change. If so, the above-mentioned first-order lag value of the throttle opening is phenomenally equal to 1 of the effective opening area.
It should correspond to the next delay almost equivalently. Therefore, FIG.
As shown in FIG. 10, the effective opening area (first order delay value) ADELAY is calculated from the first order delay value of the throttle opening (in FIG. 10, (1-B) / (z-B) is discrete. Means the first-order lag in the transfer function of the system).

That is, the throttle projection area S is obtained from the throttle opening θTH according to a preset characteristic,
A coefficient C is obtained from the throttle opening first-order delay value θTH-D and the intake pressure Pb according to the characteristics shown in FIG. 12, and then the product of the two is obtained to obtain an effective opening area (first-order delay value) ADELAY.
Was calculated. Further, the chamber filling air amount ΔG
b to eliminate the delay in reflecting the
The first-order delay of b was also used.

Further, it has been found that it is not necessary to separately obtain the throttle passing air amount Gth and the chamber filling air amount ΔGb, and the chamber filling air amount ΔGb is calculated from the throttle passing air amount Gth to obtain the cylinder intake air amount. Gc
Can be calculated only from the throttle passing air amount Gth. This has simplified the configuration and reduced the amount of calculation. That is, in equation (1), the cylinder intake air amount Gc per unit time ΔT can be expressed as in equation (5), which is equivalent to equations (6) and (7). When Expressions 6 and 7 are represented in the form of a transfer function, Expression 8 is derived. That is, as shown in Expression 8, the intake air amount Gc can be obtained from the first-order delay value of the throttle passing air amount Gth. This is shown in a block diagram in FIG. Since the transfer function in FIG. 18 is different from that in FIG. 10, a dash is given to indicate (1-B ′) / (z−B ′) in the meaning indicating the transfer function.

[0063]

(Equation 5)

[0064]

(Equation 6)

[0065]

(Equation 7)

[0066]

(Equation 8)

Therefore, the basic fuel injection amount TiM-F is calculated as follows: TiM-F = map search fuel injection amount TiM × actual throttle effective opening area / intake pressure Pb and throttle effective opening determined by the primary delay value θTH-D of the throttle opening. Area = Map search Fuel injection amount TiM × RATIO-A

Based on the above, the operation of the control device will be described with reference to the flowchart of FIG.

First, the engine speed N detected in S10
e, intake pressure Pb, throttle opening θTH, pressure Pa, engine cooling water temperature Tw, etc. are read. Note that the throttle opening θ
TH learns the throttle full-opening degree during idle operation,
A value detected based on the value is used.

Subsequently, the program proceeds to S12, in which it is determined whether or not the engine is being cranked (started). If the answer is NO, the program proceeds to S14, in which it is determined whether or not fuel cut has been performed. The engine speed Ne and the intake pressure Pb
From this, a map showing the characteristic shown in FIG. 13 stored in the ROM 72 is searched to find the fuel injection amount TiM (the fuel injection amount Timap in the steady operation state). It should be noted that pressure correction and the like are appropriately added to the obtained fuel injection amount TiM as necessary, but the correction itself is not the gist of the present invention, and therefore detailed description is omitted. Then, the program proceeds to S18, in which a primary delay value θTH-D of the detected throttle opening is calculated.

Then, the program proceeds to S22, in which the throttle opening .theta.TH
And the effective opening area A of the current throttle from the intake pressure Pb
Is calculated. Then, the program proceeds to S24, in which a first order delay value ADELAY of the effective opening area of the throttle is calculated from the throttle opening first order delay value θTH-D and the intake pressure Pb.

Then, the process proceeds to S26, where RATIO-A is calculated from the following equation: RATIO-A = (A + ABYPASS) / (A + ABYPASS) DELAY. Note that the value ABYPASS is
2 means the amount of air that is drawn into the combustion chamber without passing through the throttle valve 16 (shown as "lift amount" in FIG. 10), and this air amount is also taken into account to determine the fuel injection amount accurately. Therefore, it is necessary to convert the value corresponding to the value into the throttle opening ABYPASS according to a predetermined characteristic, add it to the effective opening area A, and sum the value (A +
ABYPASS) and its first approximation (“(A + ABYPASS) DELA
Y ”), and call it RATIO-A.

As described above, as a result of addition to both the numerator and the denominator, even if there is an error in the measurement of the amount of air taken into the combustion chamber without passing through the throttle valve, the degree of influence on the determined fuel injection amount Becomes smaller. Then, the program proceeds to S28, in which a basic fuel injection amount TiM-F corresponding to the throttle passing air amount is calculated by multiplying the fuel injection amount TiM by RATIO-A. S12
When it is determined that cranking is being performed in step S30, the routine proceeds to step S30, where a predetermined table (not shown) is retrieved from the water temperature Tw to calculate the fuel injection amount Ticr at the time of cranking. The fuel injection amount TiM-F is determined on the basis of the formula (1), and when it is determined in S14 that the fuel is cut, the routine proceeds to S34, where the fuel injection amount TiM-F is set to zero.

The above-described calculation method of the basic fuel injection amount TiM-F can express from a steady operation state to a transient operation state by a simple algorithm, and the fuel injection amount in the steady operation state is guaranteed to some extent by a map search. At the same time, the fuel injection amount can be optimally determined without requiring complicated calculations. Moreover, there is no need to change the model formula between the steady operation state and the transient operation state, and
Because all driving conditions can be expressed by two equations,
Generally, there is no control discontinuity as seen near the switching point. In addition, since the behavior of the air can be well expressed, controllability and control accuracy can be improved.

Returning to the block diagram of FIG. 8, various correction coefficients KTOTAL including an EGR correction coefficient KEGR and a canister purge correction coefficient KPUG are calculated.

First, the EGR correction coefficient will be described.

Since the exhaust gas recirculation amount becomes a disturbance when controlling the fuel injection amount of the internal combustion engine, it is necessary to accurately estimate the exhaust gas recirculation rate or the exhaust gas recirculation amount. Here, “exhaust gas recirculation rate” means a volume ratio or a weight ratio of exhaust gas / intake air.

FIG. 19 is a flow chart for explaining the operation of estimating the exhaust gas recirculation rate.

Prior to the description of the figure, an algorithm of the operation of estimating the exhaust gas recirculation rate according to the embodiment will be described with reference to FIG.

The amount of gas passing through the exhaust gas recirculation valve is determined by the opening area of the valve and the pressure ratio between before and after the valve, that is, the flow characteristics (design specifications). That is, it is considered that the value is obtained from the ratio between the opening area of the valve, that is, the lift amount and the pressure between the upstream and downstream of the valve. As shown in FIG. 20, even in an actual machine, it is considered that the amount of recirculated gas can be estimated to some extent by obtaining the lift amount of the valve and the ratio between the atmospheric pressure Pa and the intake pressure Pb of the intake pipe 12 (actually, Although the flow rate characteristic slightly changes depending on the exhaust pressure and the exhaust temperature, it is considered that the change in the characteristic can be absorbed to a considerable extent by using the gas amount ratio as described later.)

Therefore, attention was first paid to this point, and the reflux rate was determined based on the flow characteristics. Note that the opening area is obtained from the lift amount, because a valve having a structure in which the lift amount corresponds to the opening area was used. Therefore, when using another structure such as a linear solenoid, the opening area is obtained from another parameter.

The recirculation rate includes a steady-state recirculation rate and a transient recirculation rate. Of these, the steady-state recirculation rate is a value in a state where the lift command value is equal to the actual lift, and The recirculation rate is a value in a state where the lift command value is not equal to the actual lift, as shown in FIG. Then, in the algorithm according to the present invention, the difference at the time of transition is, as shown in FIG.
It was thought that it was caused by deviation from the normal reflux rate.

Specifically, in the steady state, the lift command value = actual lift, the gas amount ratio = 1. That is, the recirculation rate = the steady state recirculation rate. In the transient state, the lift command value ≠ the actual lift, the gas amount ratio ≠ 1. Reflux rate = Reflux rate at steady state (map search value) x gas amount ratio.

As described above, it was considered that the net recirculation rate flowing into the combustion chamber can be obtained by multiplying the ratio of the two gas amounts by the recirculation rate in the steady state. The expression is as follows. Net reflux rate = (steady-state reflux rate) x (gas amount QACT obtained from actual lift and pressure ratio before and after valve) / (gas amount QCMD obtained from lift command value and pressure ratio before and after valve)

Here, the recirculation rate in the steady state is obtained by obtaining a recirculation rate correction coefficient and subtracting it from 1. That is,
When the steady-state reflux rate correction coefficient is called KEGRMAP, the steady-state reflux rate = (1−KEGRMAP). In this specification, the steady state recirculation rate or the steady state recirculation rate correction coefficient is also referred to as a basic exhaust gas recirculation rate or a basic exhaust gas recirculation rate correction coefficient. Also, the steady-state recirculation rate correction coefficient KEGRMAP is determined in advance by experiments from the engine speed Ne and the intake pressure Pb, set as a map as shown in FIG. 22, and is searched for.

In the exhaust gas recirculation control, the lift command value of the exhaust gas recirculation valve is determined from the engine speed and the engine load, etc., as shown in FIG. Value) has a delay. Further, there is a delay in the flow of the recirculated gas into the combustion chamber in accordance with the valve opening operation.

Therefore, the applicant of the present application has previously filed Japanese Patent Application No.
No. 0,557, the above equation, net reflux rate = (reflux rate at steady state) × (gas amount QACT obtained from actual lift and pressure ratio before and after valve) / (calculated from lift command value and pressure ratio before and after valve. Gas amount QCMD), the method of calculating the net recirculation rate was described, but the concept of the first-order delay was used for the inflow delay of the recirculation gas. Here, using the concept of dead time,
The recirculated gas that has passed through the exhaust gas recirculation valve can be regarded as flowing into the combustion chamber at one time after a certain dead time has elapsed. Therefore, the above-described net recirculation rate is calculated for each predetermined cycle and stored in the storage means, and the calculated value of the past cycle corresponding to the dead time is used to calculate the recirculation rate of the exhaust gas truly flowing into the combustion chamber. I considered it.

Hereinafter, the operation of the apparatus according to the embodiment will be described with reference to FIG.
The description will be made according to the 9 flow chart. This program is started at each TDC position.

First, at S200, the engine speed Ne, the intake pressure Pb, the atmospheric pressure Pa, the actual lift LACT (the lift sensor 1
23, and the like, and the program proceeds to S202, in which a lift command value LCMD is retrieved from the engine speed Ne and the intake pressure Pb. Here, the lift command value LCMD is, as shown in FIG.
A map is determined by setting characteristics in advance, and the map is determined.

Then, the program proceeds to S204, in which a map shown in FIG. 22 is retrieved from the engine speed Ne and the intake pressure Pb to obtain a basic exhaust gas recirculation rate correction coefficient KEGRMAP. Next, proceeding to S206, it is confirmed that the detected actual lift LACT is not zero, that is, it is confirmed that the exhaust gas recirculation valve 122 is opened, and the program proceeds to S208, where the retrieved lift command value LC
MD is compared with a predetermined lower limit value LCMDLL (small value).

If it is determined in S208 that the search value is not less than the lower limit, the process proceeds to S210, where the intake pressure Pb
A ratio Pb / Pa between the pressure and the atmospheric pressure Pa is obtained, and a map (not shown) of the characteristic shown in FIG. 20 is retrieved from the retrieved lift command value LCMD to obtain a gas amount QCMD.
This is the "gas amount obtained from the lift command value and the pressure ratio before and after the valve" in the above formula.

Subsequently, the flow advances to S212, where the detected actual lift is detected.
A map (not shown) of the characteristic shown in FIG. 20 is similarly retrieved from the ratio Pb / Pa similar to LACT, and the gas amount QA
Find the CT. This is equivalent to the "gas amount obtained from the pressure ratio between the actual lift and the pressure before and after the valve" in the above formula.

Subsequently, the program proceeds to S214, in which a value obtained by subtracting the retrieved basic exhaust gas recirculation rate correction coefficient KEGRMAP from 1 is defined as a steady-state recirculation rate (basic exhaust gas recirculation rate or steady-state recirculation rate). Here, the steady-state recirculation rate is, as described above, a recirculation rate when the exhaust gas recirculation operation is stable, that is, is not in a transient state such as when the exhaust gas recirculation operation is started or stopped. Means the reflux rate.

Then, the program proceeds to S216, where the net reflux rate is determined by multiplying the steady-state reflux rate by the ratio QACT / QCMD of the values QACT and QCMD, as shown in the figure. Then, the program proceeds to S218, in which a fuel injection correction coefficient KEGRN for the exhaust gas recirculation rate is calculated. FIG. 24 is a subroutine flowchart showing the operation.

Referring to the figure, in S300, the net recirculation rate (determined in S216 in FIG. 19) is subtracted from 1, and the value is used as a fuel injection correction coefficient KEGRN for the exhaust recirculation rate. Then, the process proceeds to S302, in which the fuel injection correction coefficient KEGRN for the calculated exhaust gas recirculation rate is stored (stored) in the ring buffer. FIG. 25 is an explanatory diagram showing the configuration of the ring buffer.
4. As shown, the ring buffer has n addresses, and each address is numbered from 0 to n. Each time the fuel injection correction coefficient KEGRN is calculated by starting the flowchart of FIG. 19 (and FIG. 24) at TDC, the fuel injection correction coefficient KEGRN is sequentially stored (updated) from the top in the figure.

Then, the program proceeds to S304, in which a map is retrieved from the detected engine speed Ne and the engine load, for example, the intake pressure Pb, to find a dead time τ. FIG. 26 is an explanatory diagram showing the characteristics. That is, the above-mentioned dead time indicates a delay time until the recirculated gas passing through the exhaust gas recirculation valve flows into the combustion chamber, and varies depending on the engine speed and the engine load, for example, the intake pressure. Here, the dead time τ is more specifically indicated by the buffer number described above.

Subsequently, the flow proceeds to S306, where the calculated value (the fuel injection correction coefficient KEGRN for the exhaust gas recirculation rate) stored at the corresponding address is read based on the found dead time τ (more specifically, the buffer number). That is, as shown in FIG. 27, when the current time point is A, for example, the calculated value 12 times before is selected, and is set as the fuel injection correction coefficient KEGRN for the current exhaust gas recirculation rate.

When this is viewed from the operation of the exhaust gas recirculation valve, 12
The fuel injection correction coefficient KEGRN for the previous exhaust gas recirculation rate is 1.0, which means that the exhaust gas recirculation valve has been closed. Thereafter, the fuel injection correction coefficient KEGRN with respect to the exhaust gas recirculation rate gradually decreases, for example, to 0.99, 0.98, etc. In other words, the exhaust gas recirculation valve is opened to the present time point A. At the moment,
It is determined that the recirculated gas has not yet flowed into the combustion chamber, so that the reduction correction of the fuel injection is not performed.

At the same time, the fuel injection amount is corrected based on the fuel injection correction coefficient KEGRN for the determined exhaust gas recirculation rate.
This fuel injection amount is corrected by multiplying a basic fuel injection amount TiM-F obtained from the engine speed and the engine load by a fuel injection correction coefficient KEGRN for the exhaust gas recirculation rate, which will be described later.
This is done by finding l.

Incidentally, in the flow chart of FIG.
When the actual lift LACT is determined to be zero in 206, the exhaust gas recirculation is not performed, but the fuel injection correction coefficient KEGRN for the exhaust gas recirculation rate is determined from the value after the dead time τ has elapsed. Proceeding to S214 and thereafter, a fuel injection correction coefficient KEGRN for the net recirculation rate and the exhaust recirculation rate is calculated. In this case, the net reflux rate becomes 0 in S216, and FIG.
In S300 of the flow chart, the fuel injection correction coefficient KEGRN for the exhaust gas recirculation rate is determined to be 1.0.

If it is determined in step S208 that the lift command value LCMD is equal to or smaller than the lower limit value LCMDLL, the process proceeds to step S222, and the lift command value LCMD retains the previous value LCMDk-1 (for simplicity, in this specification, This time we omit adding k to the value).

This is because when the shift from the region in which the exhaust gas recirculation is performed to the region in which the exhaust gas recirculation is not performed, even if the lift command value LCMD becomes zero, there is a delay in the dynamic characteristics of the exhaust gas recirculation valve 122. Since the lift command value LCMD does not become zero, when the lift command value LCMD is equal to or lower than the lower limit (threshold) LCMDLL, the lift command value LCMD is held at the previous value LCMDk-1 (the value at the time of the previous control cycle k-1). did. This previous value hold is performed until it is confirmed in S206 that the actual lift LACT has become zero.

When the lift command value LCMD is equal to or smaller than the lower limit value LCMDLL, the lift command value LCMD may be zero. In this case, the QCMD search value in S210 becomes zero and S216
Is divided by zero and the calculation becomes impossible. However, by holding the previous value as described above, there is no possibility that the calculation becomes impossible. Although the lower limit value LCMDLL is a minute value, it may be zero.

Then, the process proceeds to S224, in which the map search value (searched in S204) of the basic exhaust gas recirculation rate correction coefficient KEGRMAP is replaced with the previous search value KEGRMAPk-1. This is because, in the operating state in which the lift command value LCMD searched in S202 is determined to be equal to or less than the lower limit value, the basic exhaust gas recirculation rate correction coefficient KEGRMAP searched in S204 is set to 1 in the characteristic expected in this embodiment. This is because there is a possibility that the steady-state reflux rate becomes zero in the calculation of S214.

As described above, the net recirculation rate of the exhaust gas flowing into the combustion chamber through the exhaust gas recirculation valve is calculated from the detected engine speed and the engine load, for example, the intake pressure and the operating state of the exhaust gas recirculation valve. It is calculated for each cycle, and based on it, the fuel injection correction coefficient for the exhaust gas recirculation rate is sequentially calculated and stored for each calculation cycle, and the waste gas until the exhaust gas passes through the exhaust gas recirculation valve and flows into the combustion chamber. Ask for time,
Since the calculation value of the calculation cycle corresponding to the dead time is selected and is regarded as the fuel injection correction coefficient for the exhaust gas recirculation rate in the current calculation cycle, complicated calculations and uncertain calculation elements are reduced as much as possible. This makes it possible to accurately determine the recirculation rate of the exhaust gas flowing into the combustion chamber and accurately correct the fuel injection amount, with a simple configuration. In the above description, the net reflux rate may be stored in the ring buffer instead of KEGRN, and the dead time τ may be a fixed value. The details are described in Japanese Patent Application No. 6-294,014, which was previously proposed by the present applicant, and further description will be omitted.

Next, the canister purge correction coefficient KPUG
(According to the purge mass) will be described.

When the canister is purged, the canister 2
Since the gas containing fuel is sucked into the intake system from 23, the air-fuel ratio shifts to the rich side. This shift is corrected later in the feedback system. However, at the time of canister purging, since the air-fuel ratio is expected to shift to the rich side in advance, if the amount of correction for reduction corresponding to the purge mass is corrected in advance as KPUG, the correction amount of the feedback system decreases, that is, Since the load on the feedback system is reduced,
Stability and followability with respect to disturbance are improved.

As the correction method, the inflow canister
A method of calculating the fuel amount during canister purging from the purge flow rate and the concentration, or a method of obtaining a correction coefficient KPUG according to the purge mass from a deviation of the air-fuel ratio sensor from the target air-fuel ratio is considered. An example of calculating the canister purge correction coefficient KPUG based on the former method will be described below.

FIG. 28 is a flow chart showing the calculation method.

First, in step S400, the flow rate of the canister purge is detected via the flow meter 226, and in step S402, the concentration is detected via the HC concentration sensor 227. Next, the inflow fuel amount (mass) by the canister purge is calculated from the flow rate and the concentration detected in S404. Next, in S406, the calculated inflow fuel amount is converted into a gasoline fuel amount. That is, most of the fuel component in the canister purge is butane, which is a light component of gasoline.
Since the stoichiometric air-fuel ratio is different between butane and gasoline, it is converted to gasoline equivalent here. Then, proceed to S408,
A cylinder intake air amount Gc is obtained by multiplying the map search fuel injection amount TiM by the target air-fuel ratio, and a correction coefficient KPUG corresponding to a purge mass is calculated from the cylinder intake air amount Gc and the converted gasoline amount.

The control of the purge control valve 225 is performed by a program (not shown) so as to satisfy the target canister purge amount in accordance with a predetermined operating state such as an engine speed and an engine load. Needless to say, when the canister purging is not performed, the correction coefficient KPUG corresponding to the purge mass becomes 1.

In the above description, a correction coefficient KPUG corresponding to the target purge mass, for example, 0.95 may be set first, and the purge control valve may be controlled to match the value. Further, as described above, the correction coefficient KPUG according to the purge mass may be obtained from the deviation of the air-fuel ratio sensor from the target air-fuel ratio. Further, the cylinder intake air amount Gc may be set as a map value from the engine speed and the engine load. Further, S40
The gasoline fuel amount obtained in step 6 may be subtracted from the required fuel injection amount Tcyl.

In addition, the correction coefficient KTOTAL includes a correction coefficient based on the water temperature and a correction coefficient based on the intake air temperature. Fuel injection correction coefficient KEGRN for exhaust gas recirculation rate thus obtained, KP according to purge mass
The basic fuel injection amount TiM-F is multiplied as KTOTAL by adding UG and the like, and is corrected.

Next, a target air-fuel ratio KCMD and a target air-fuel ratio correction coefficient KCMDM are calculated.

FIG. 29 is a flow chart showing the calculation operation.

First, at S500, the basic value KB
Search for S. This is obtained by searching the map shown in FIG. 14 from the engine speed Ne and the intake pressure Pb. It should be noted that the map includes a basic value at the time of idling. Further, in the case of a so-called lean burn engine in which the air-fuel ratio supplied to the engine at a low load of the engine is increased (smaller in terms of the equivalent ratio) to improve the fuel efficiency characteristics, a lean burn basic value is also included.

Then, the program proceeds to S502, in which it is determined whether or not the lean burn control after starting the engine is being executed by referring to the value of an appropriate timer. Since the internal combustion engine 10 according to the embodiment is provided with a variable valve timing mechanism, the target air-fuel ratio is set slightly leaner than the stoichiometric air-fuel ratio for a predetermined period after the start by stopping one operation of the intake valve. Lean burn control is set to the side. That is, by enriching the air-fuel ratio while the catalyst device after the start is not yet activated, the problem of increasing HC is avoided.

In a normal engine having two intake valves, if the target air-fuel ratio is set to the lean side after starting the engine, combustion of the engine becomes unstable, and a misfire may occur. However, in the engine equipped with the variable valve timing mechanism according to the embodiment, when one of the intake valves is stopped, a swirl called so-called swirl is formed in the intake air in the combustion chamber. However, since stable combustion can be obtained, leaning can be achieved even immediately after starting. Therefore, it is determined from the timer value whether or not it is in the period,
The lean correction coefficient is calculated accordingly. This value is calculated to be, for example, 0.89 if in the lean burn control period and 1.0 if not.

Next, the program proceeds to S504, in which it is determined whether or not the throttle opening is fully open (WOT), and a fully-open increase correction value is calculated according to the result of the determination. Next, in S506, it is determined whether or not the water temperature Tw is high, and an increase correction coefficient KTWOT is calculated according to the determination result. This value includes a correction coefficient for protecting the engine at a high water temperature.

Then, the program proceeds to S508, in which the basic value KBS is multiplied by the obtained correction coefficient to correct the basic value KBS, and the target air-fuel ratio KCMD is determined. This is the corrected base value KBS
As shown in FIG. 7, in the range where the output of the O ₂ sensor 56 near the stoichiometric air-fuel ratio has a linear characteristic (shown by a broken line on the vertical axis), the minute control of the air-fuel ratio (the above-described MIDO ₂
The control is performed by setting a window (hereinafter referred to as DKCMD-OFFSET) for the control and adding the window value DKCMD-OFFSET to the corrected basic value KBS. That is, the target air-fuel ratio
KCMD is determined as follows. KCMD = KBS + DKCMD-OFFSET

Then, the program proceeds to S510, in which the target air-fuel ratio is determined.
Performs KCMD (k) (k: time) limit processing. Then S
The process proceeds to 512, where it is determined whether the calculated target air-fuel ratio KCMD (k) is 1 or a value in the vicinity thereof.
Proceed to 514 to activate determination of the O ₂ sensor 56. This is executed by another routine (not shown), and the O ₂ sensor 56
The detection is performed by detecting a change in the output voltage. Then S51
Proceed to 6 to perform DKCMD calculation for MIDO ₂ control. This is downstream of the first catalyst device 28 (the catalyst device 28 shown in FIG. 5).
In the case of ( ₂ ), the target air-fuel ratio KCMD (k) of the LAF sensor 54 on the upstream side of the output of the O ₂ sensor 56 (downstream of the first CAT bed)
Means a variable operation. For more information as shown in FIG. 7, performed by calculating a value DKCMD using the PID control law in the deviation of the output voltage VO ₂ M of the predetermined comparison voltage VrefM and the O ₂ sensor 56. The comparison voltage VrefM is obtained according to the atmospheric pressure Pa, the water temperature Tw, the exhaust volume (which can be obtained from the engine speed Ne and the intake pressure Pb), and the like.

The above window value DKCMD-OFFSET
Is an offset value added by the first and second catalyst devices 28 and 30 to maintain an optimum purification rate. Since this depends on the characteristics of the catalyst device, it is determined in consideration of the characteristics of the first catalyst device 28 in the illustrated example. In addition, since it changes due to aging, learning is performed by a weighted average using the calculated value of the value DKCMD every time. Specifically, DKCMD-OFFSET (k) = W × DKCMD + (1-W) × DKCMD-OF
Obtained by FSET (k-1). Here, W: weighting factor, k: time. That is, by learning and calculating the target air-fuel ratio KCMD using the previous value of the value DKCMD-OFFSET, it is possible to perform feedback control to the air-fuel ratio at which the purification rate becomes optimal without being affected by aging. This learning may be performed by dividing the operating state for each region from the engine speed Ne and the intake pressure Pb.

Then, the flow advances to S518, where the calculated value DKCMD is calculated.
(k) is added to update the target air-fuel ratio KCMD (k), and S520
The table showing the characteristics shown in FIG. 30 is stored in the target air-fuel ratio.
Search with KCMD (k) to find the correction coefficient KETC. This is to compensate for the difference in the charging efficiency of the intake air due to the heat of vaporization. Specifically, using the obtained correction coefficient KETC, KCMD
(k) is corrected as shown in the figure, and the target air-fuel ratio correction coefficient KCMDM (k)
Is calculated. That is, in this control, the target air-fuel ratio is indicated by the equivalent ratio, and a value KCMD obtained by performing a charging efficiency correction on the target air-fuel ratio is indicated.
Let M be the target air-fuel ratio correction coefficient. When the result in S512 is NO, the target air-fuel ratio KCMD to be controlled is greatly deviated from the stoichiometric air-fuel ratio. For example, during lean burn operation, there is no need to perform MIDO ₂ control. Jumps to S520 immediately. Finally S522
Then, the limit processing of the target air-fuel ratio correction coefficient KCMDM (k) is performed, and the processing ends.

As shown in the block diagram of FIG. 8, the target air-fuel ratio correction coefficient KCMDM and the total value KTOTAL of the various correction coefficients thus obtained are multiplied by the basic fuel injection amount TiM-F to obtain the required fuel injection amount Tcy.
l is calculated.

Subsequently, a feedback correction coefficient such as KSTR is calculated. Before starting the description, the sampling of the LAF sensor output and the observer will be described. The sampling operation block is shown in FIG.
lV ”.

Since the exhaust gas is discharged in the exhaust stroke in the internal combustion engine, the behavior of the air-fuel ratio in the exhaust system assembly of the multi-cylinder internal combustion engine is clearly synchronized with TDC. Therefore, when sampling the air-fuel ratio by providing the LAF sensor 54 in the exhaust system of the internal combustion engine, it is necessary to perform the sampling in synchronization with TDC. However, depending on the sampling timing of the control unit (ECU) 34 that processes the detection output, the air-fuel ratio In some cases, the behavior of the object cannot be accurately grasped. That is, for example, when the air-fuel ratio of the exhaust system collecting part with respect to TDC is as shown in FIG. 31, the air-fuel ratio recognized by the control unit is as shown in FIG.
As shown in (1), the value is completely different depending on the sample timing. In this case, it is desirable to perform sampling at a position where the actual change in the output of the air-fuel ratio sensor can be grasped as accurately as possible.

Furthermore, the change in the air-fuel ratio also differs depending on the exhaust gas arrival time at the sensor and the reaction time of the sensor.
The arrival time at the sensor changes depending on the exhaust gas pressure, the exhaust gas volume, and the like. Further, since sampling is performed in synchronization with TDC based on the crank angle, it is inevitably affected by the engine speed. As described above, the detection of the air-fuel ratio largely depends on the operating state of the engine. For this purpose, in the prior art, for example, the technology described in JP-A-1-313,644, it is determined whether the detection is appropriate or not at every predetermined crank angle. In such a case, there is a possibility that the air conditioner may not be able to cope with the problem, and at the time when the detection is determined, the inflection point of the output of the air-fuel ratio sensor may be missed.

FIG. 33 is a flow chart showing the sampling operation of the LAF sensor. Since the detection accuracy of the air-fuel ratio is closely related to the estimation accuracy of the observer in particular, the description will be made before the description of FIG. Here, the air-fuel ratio estimation by the observer will be briefly described.

First, in order to accurately separate and extract the air-fuel ratio of each cylinder from the output of one LAF sensor, it is necessary to accurately clarify the detection response delay of the LAF sensor. Therefore, this delay is simulated as a first-order delay system, and FIG.
A model as shown in FIG. Where LAF: LAF
Assuming that the sensor output and A / F are input A / F, the state equation can be expressed by the following equation (9).

[0130]

(Equation 9)

When this is discretized by the period ΔT, it becomes as shown in Expression 10. FIG. 35 shows Expression 10 in a block diagram.

[0132]

(Equation 10)

Therefore, the true air-fuel ratio can be obtained from the sensor output by using equation (10). That is, if Equation 10 is transformed, Equation 11 is obtained.
The value at time k-1 can be inversely calculated from the value at time as shown in Expression 12.

[0134]

[Equation 11]

[0135]

(Equation 12)

More specifically, if Equation 10 is expressed as a transfer function using Z-transformation, Equation 13 is obtained. Therefore, the previous input air-fuel ratio can be calculated in real time by multiplying the inverse transfer function by the current LAF sensor output LAF. Can be estimated. FIG. 36 shows a block diagram of the real-time A / F estimator.

[0137]

(Equation 13)

Next, a method of separating and extracting the air-fuel ratio of each cylinder based on the true air-fuel ratio obtained as described above will be described. The value at the time k is considered as a weighted average considering the temporal contribution of the air-fuel ratio of the cylinder, and the value at the time k is expressed as in Expression 14. Note that "F / A" is used here because F (fuel amount) is a control amount, but "Air / fuel ratio" will be used in the following description for convenience of understanding unless there is a problem. Note that the air-fuel ratio (or the fuel-air ratio) means a true value obtained by correcting the response delay previously obtained by Expression 13.

[0139]

[Equation 14]

That is, the air-fuel ratio of the collecting portion is determined by adding a weight Cn to the past combustion history of each cylinder (for example, the most recently burned cylinder is 4%).
0%, before that 30%. . . , Etc.). This model is represented by a block diagram as shown in FIG.

The state equation is as shown in Equation 15.

[0142]

(Equation 15)

If the air-fuel ratio of the collecting portion is set to y (k), the output equation can be expressed as in equation (16).

[0144]

(Equation 16)

In the above, since u (k) cannot be observed, even if an observer is designed from this equation of state, x (k)
Cannot be observed. So 4 TDC before (ie,
Assuming that the air-fuel ratio of the same cylinder is in a steady operation state in which the air-fuel ratio does not suddenly change, x (k + 1) = x (k-3),
become that way.

[0146]

[Equation 17]

Here, simulation results are shown for the model obtained as described above. FIG. 38 shows that in the four-cylinder internal combustion engine, the air-fuel ratio of three cylinders is set to 14.7,
The case where the fuel is supplied at 2.0 is shown. FIG. 39 shows the air-fuel ratio of the collecting portion at that time obtained by the above model. In the figure, a step-like output is obtained. Here, considering the response delay of the LAF sensor,
The sensor output has a waveform as shown as "model output value" in FIG. "Actual measured value" in the figure is LA for the same case
The measured value of the output of the F sensor is compared with this value, and it is verified that the above-described model models the exhaust system of the multi-cylinder internal combustion engine well.

Therefore, the problem is reduced to the problem of the ordinary Kalman filter for observing x (k) in the state equation and the output equation expressed by the equation (18). When the Riccati equation is solved with the weight matrices Q and R as shown in Equation 19, the gain matrix K becomes as shown in Equation 20.

[0149]

(Equation 18)

[0150]

[Equation 19]

[0151]

(Equation 20)

From this, A-KC is obtained as shown in Equation 21.

[0153]

(Equation 21)

The general observer configuration is as shown in FIG. 41. However, since there is no input u (k) in this model, as shown in FIG. 42, the configuration is such that only y (k) is input. When this is expressed by a mathematical expression, it becomes as shown in Expression 22.

[0155]

(Equation 22)

Here, an observer that receives y (k) as an input, that is, a system matrix of a Kalman filter is represented by Expression 23.

[0157]

(Equation 23)

In the present model, when the elements of the weight distribution R of the Riccati equation: the elements of Q = 1: 1, the system matrix S of the Kalman filter is given by Expression 24.

[0159]

(Equation 24)

FIG. 43 shows a combination of the above-described model and observer. The simulation result is omitted since it is shown in the earlier application, but the air-fuel ratio of each cylinder can be accurately extracted from the air-fuel ratio of the collecting portion.

Since the air-fuel ratio of each cylinder can be estimated from the air-fuel ratio of the collecting section by the observer, the air-fuel ratio can be controlled for each cylinder using a control law such as PID. Specifically, as shown in FIG. 44, the PID control law is used to collect the PID control rules from the sensor output (A / F, ie, the detected air-fuel ratio KACT) and the past value of the cylinder-by-cylinder feedback correction coefficient. The feedback correction coefficient KLAF is obtained, and the observer estimates # nA / F for each cylinder.
, A feedback correction coefficient #nKLAF (n: cylinder) for each cylinder is obtained. Feedback correction coefficient #nKL for each cylinder
More specifically, AF divides the collective portion A / F, that is, KACT, by the previous calculated value of the average value of all cylinders of the feedback correction coefficient #nKLAF for each cylinder (using a division symbol instead of an addition point). The PI is determined so as to eliminate the deviation between the target value obtained as described above and the observer estimated value # nA / F.
Determined using D-rule.

As a result, the air-fuel ratio of each cylinder converges to the collecting air-fuel ratio, and the collecting air-fuel ratio converges to the target air-fuel ratio. Converge. Here, the fuel injection amount #nTout of each cylinder (defined by the valve opening time of the injector) is obtained by: # nTout = Tcyl × # nKLAF × KLAF (n: cylinder). Since the details of such control are described in Japanese Patent Application No. 5-251138 previously proposed by the present applicant, further description will be omitted.

Here, returning to the flow chart of FIG. 33, the sampling of the LAF sensor output will be described. This program is started at the TDC position.

This will be described below with reference to the flowchart shown in FIG. First, in S600, the engine speed Ne, the intake pressure Pb, and the valve timing V / T are read out.
Proceeding to S606, a timing map (described later) for HiV / T or LoV / T is searched, and proceeding to S608, HiV / T
Sampling of sensor output used for observer operation for LoV / T and LoV / T is performed. Specifically, a timing map is searched from the engine speed Ne and the intake pressure Pb, and one of the twelve buffers described above is assigned to its No. And select the sampling value stored therein.

FIG. 45 is an explanatory diagram showing the characteristics of the timing map. As shown in the drawing, the characteristics are such that a value sampled at a faster crank angle is selected as the engine speed Ne is lower or the intake pressure (load) Pb is higher. Is set to Here, “early” means a value sampled at a position closer to the previous TDC position (in other words, an old value). Conversely, the higher the engine speed Ne or the lower the intake pressure Pb, the slower the crank angle, that is, the later TDC
A setting is made so that a value sampled at a crank angle close to the position (in other words, a new value) is selected.

That is, as shown in FIG. 32, it is best to sample the output of the LAF sensor at a position as close as possible to the inflection point of the actual air-fuel ratio. Assuming that the reaction time of the sensor is constant, the value occurs at a faster crank angle as the engine speed decreases, as shown in FIG. Also, it is expected that as the load increases, the exhaust gas pressure and the exhaust gas volume increase, and accordingly, the flow rate of the exhaust gas increases and the arrival time at the sensor is shortened. In that sense, the sample timing is shown in FIG.
The settings were as shown in FIG.

Further, regarding the valve timing, an arbitrary value Ne1 of the engine speed is changed to Ne1-Lo, Hi for the Lo side.
Ne1−Hi for the side and Pb1-Lo for the Lo side and Pb1-H for the Hi side for the intake pressure.
Assuming that i, the map characteristics are Pb1-Lo> Pb1-Hi Ne1-Lo> Ne1-Hi. That is, in the case of HiV / T, the opening time of the exhaust valve is earlier than that of LoV / T, so that if the values of the engine speed or the intake pressure are the same, an earlier sampling value is selected. Is set.

Then, the flow advances to S610 to perform the operation of the observer matrix on HiV / T, and then to S612, the same operation is performed on LoV / T. Subsequently, the process proceeds to S614, where the valve timing is determined again, and the process proceeds to S616, S618 according to the determination result, and the calculation result is selected and terminated.

In other words, the behavior of the air-fuel ratio collecting part changes with the switching of the valve timing, so that it is necessary to change the observer matrix. However, the estimation of the air-fuel ratio of each cylinder is not instantaneous, and the calculation of the air-fuel ratio of each cylinder requires several operations before the convergence is completed. The calculation using the changed observer matrix is performed in an overlapping manner, so that even if the valve timing is changed, it can be selected according to the changed valve timing in S614. After each cylinder is estimated, as described above, a feedback correction coefficient is obtained so as to eliminate the deviation from the target value, and the injection amount is determined.

With this configuration, the detection accuracy of the air-fuel ratio can be improved. That is, as shown in FIG. 47, since sampling is performed at relatively short intervals, the sampled value almost exactly reflects the sensor output, and the values sampled at relatively short intervals are sequentially stored in the buffer group. Further, the inflection point of the sensor output is predicted in accordance with the engine speed and the intake pressure (load), and a value corresponding thereto is selected from a group of buffers at a predetermined crank angle. Thereafter, an observer calculation is performed to estimate the air-fuel ratio of each cylinder, and as described with reference to FIG. 44, feedback control of the air-fuel ratio for each cylinder is performed.

Therefore, as shown in the lower part of FIG.
The core 70 can accurately recognize the maximum value and the minimum value of the sensor output. Therefore, even when estimating the air-fuel ratio of each cylinder using the above-described observer with this configuration, a value approximating the actual behavior of the air-fuel ratio can be used, and the estimation accuracy of the observer is improved. The accuracy in performing the cylinder-by-cylinder air-fuel ratio feedback control described with respect to 44 is also improved.

Incidentally, regarding the sensor output sampling,
Instead of actually determining which characteristic the valve timing is in, the characteristic may be determined for both Lo and Hi characteristics, and then the characteristic may be determined only after that. LAF
The reaction time of the sensor is shorter when the air-fuel ratio of the mixture to be detected by the sensor is lean than when the air-fuel ratio is rich. It is desirable to select the sampling value detected in. Also, when a vehicle equipped with an internal combustion engine travels at high altitude, since the atmospheric pressure decreases and the exhaust pressure decreases, the arrival time of the exhaust gas to the sensor is shorter than in low altitude, It is desirable to select a sampling value detected at an earlier crank angle as the altitude increases. Further, when the LAF sensor is deteriorated, the responsiveness is reduced and the reaction time is prolonged. Therefore, as the degree of deterioration increases, it is desirable to select a sampling value detected at a later crank angle. However, the details are
Since it is described in detail in Japanese Patent Application No. 6-243277 previously proposed by the present applicant, further description is omitted.

Next, the calculation of the feedback correction coefficient such as KSTR will be described.

In the air-fuel ratio control of the internal combustion engine, FIG.
As shown in the above, a PID controller is generally used,
The feedback correction coefficient is obtained by multiplying the deviation between the target value and the manipulated variable (output of the controlled object) by the proportional term, the integral term, and the derivative term. Recently, the feedback correction coefficient can be obtained using modern control theory. Proposed.

As described above, in this application also, in the MIDO ₂ control, there is a response delay of the engine only by multiplying the fuel injection amount calculated in the feedforward system by the target air-fuel ratio correction coefficient KCMDM. Therefore, the target air-fuel ratio KCMD becomes an animated detected air-fuel ratio KACT, so that the response of the collective portion feedback correction shown in FIG. 44 is intended to dynamically compensate the response of the detected air-fuel ratio KACT from the target air-fuel ratio KCMD. Instead of the coefficient KLAF, a feedback correction coefficient KSTR is obtained by using the adaptive controller STR, and is multiplied by the fuel injection amount calculated by the feedforward system.

When the feedback correction coefficient is determined using modern control theory as in an adaptive controller, the control amount oscillates depending on the operating state because the response of the control is relatively high. May decrease. In a predetermined driving state such as during a cruise of the vehicle, the fuel supply is stopped (fuel cut), and as shown in FIG. 48, the air-fuel ratio is controlled in an open loop (O / L) during the fuel cut.

Then, when the fuel supply is restarted to reach, for example, the stoichiometric air-fuel ratio, the fuel supply amount is determined and supplied by the feedforward system according to the characteristics obtained in advance through experiments. As a result, the true air-fuel ratio suddenly changes from lean to 14.7. However, it takes some time for the supplied fuel to burn and reach the position where the air-fuel ratio sensor is disposed, and the air-fuel ratio sensor itself has a detection delay. for that reason,
The detected air-fuel ratio does not match the actual air-fuel ratio, but becomes a value shown by a broken line in FIG.

At this time, when the feedback correction coefficient is determined based on the adaptive control law, the adaptive controller STR determines the gain KSTR so as to eliminate the deviation between the target value and the detected value at once. However, this difference is due to a sensor detection delay or the like, and the detected value does not indicate the true air-fuel ratio. Nevertheless, since the adaptive controller tries to absorb this relatively large difference at once, as shown in FIG. 48, the KSTR oscillates greatly, and as a result, the control amount also oscillates and the control stability decreases. I do.

[0179] Such inconvenience does not occur only when returning from the fuel cut. The same applies when returning from the full-open increase control to the feedback control, or when returning from the lean burn control to the stoichiometric air-fuel ratio control. Furthermore, the same applies when switching from perturbation control for intentionally oscillating the target air-fuel ratio to control for a constant target air-fuel ratio. In other words, this is a common problem when the target air-fuel ratio fluctuates greatly.

Therefore, it is desirable to determine the feedback correction coefficient using the adaptive control law, the PID control law, and the like, and to switch appropriately according to the operating state. However, when switching the feedback correction coefficients determined based on different control laws, since the characteristics are different from each other, a step occurs in the correction coefficients, the operation amount changes suddenly, the control amount becomes unstable, and the control amount becomes unstable. The stability may be reduced.

Therefore, in the embodiment, the feedback correction coefficient is determined by using the adaptive control law, the PID control law, and the like, and the switching is appropriately performed according to the operating state, and the switching is performed smoothly. This prevents the control amount from becoming unstable due to the sudden change in the manipulated variable, thereby preventing the control stability from deteriorating.

FIG. 49 is a flow chart showing the operation of calculating the KSTR and the like. For convenience of understanding, the aforementioned adaptive controller STR will be described with reference to FIG. More specifically, the adaptive controller includes a STR controller (STR CONTROLLER) and an adaptive parameter adjusting mechanism (also referred to as a "parameter adjusting mechanism") as shown in the figure.

As described above, first, the required fuel injection amount Tcyl is calculated in the feedforward system, and based on the calculated required fuel injection amount Tcyl, the output fuel injection amount Tout is determined as described later. The fuel is sent to the engine 10) via the fuel injection valve 22. Target air-fuel ratio KCMD (k) of feedback system and control amount (detected air-fuel ratio) KACT (k)
(Control plant output y (k)) is input to the STR controller, and the STR controller calculates a feedback correction coefficient KSTR (k) using a recurrence formula. That is, the STR controller receives the coefficient vector θ hat (k) identified by the parameter adjusting mechanism and forms a feedback compensator.

One of the adjustment rules (mechanisms) of adaptive control is I.
D. There is a parameter adjustment rule proposed by Landau et al. This method converts the adaptive control system into an equivalent feedback system consisting of linear and nonlinear blocks, and adjusts the nonlinear blocks so that Popov's integral inequality for input and output is satisfied and the linear blocks are strongly positive. This is a method that guarantees the stability of an adaptive control system by determining a rule. In other words, in the parameter adjustment rule proposed by Landau et al., The adjustment rule (adaptive rule) expressed in a recurrence form is stabilized by using at least one of the above-mentioned Popov's hyperstability theory or Lyapunov's direct method. Is guaranteed.

This method is described in, for example, “Computer Roll” (Corona) No. 27, 28-41, or "Automatic Control Handbook" (Ohm), pages 703-707, "A Survey of Model Reference Adaptive
Techniques-Theory and Ap-plication "ID LANDAU
"Automatica" Vol. 10, pp. 353-379, 1974, "Unifi-c
ation of Discrete Time Explicit Model Reference A
daptive ControlDesigns "IDLANDAU and others" Automatic
a "Vol. 17, No. 4, pp. 593-611, 1981, and" Comb
ining Model Reference Adaptive Controllers and Sto
chasticSelf-tuning Regulators "ID LANDAU" Autom
atica ", Vol. 18, No. 1, pp. 77-84, 1982.

In the adaptive control technique of the illustrated example, the adjustment law of Landau et al. Is used. To explain below, Landau et al.'S adjustment rule states that the transfer function B (Z ^-1 ) / A
When the polynomial of the denominator and numerator of (Z ^-1 ) is expressed as in Equations 25 and 26, the adaptive parameter θ hat (k) identified by the parameter adjustment mechanism is represented by a vector (transposed vector) as shown in Equation 27. . Further, the input ζ (k) to the parameter adjustment mechanism is determined as in Expression 28. Here, m
= 1, n = 1, d = 3, that is, a plant having a dead time of three control cycles in the primary system is taken as an example.

[0187]

(Equation 25)

[0188]

(Equation 26)

[0189]

[Equation 27]

[0190]

[Equation 28]

Here, the adaptive parameter θ hat shown in Expression 27 is a scalar quantity b0 hat for determining the gain.
^-1 (k), the control element BR hat expressed using the manipulated variable
Control element S expressed using (Z ⁻¹ , k) and a control amount
(Z ⁻¹ , k), and are expressed as Equations 29 to 31, respectively.

[0192]

(Equation 29)

[0193]

[Equation 30]

[0194]

(Equation 31)

The parameter adjustment mechanism identifies and estimates the scalar amount and each coefficient of the control element, and sends them to the STR controller as the adaptive parameter θ hat shown in the above equation (26). The parameter adjustment mechanism controls the operation amount u of the plant.
(I) and control amount y (j) (i and j include past values)
Is used to calculate the adaptive parameter θ hat so that the deviation between the target value and the control amount becomes zero. The adaptive parameter θ hat is specifically calculated as shown in Expression 32. Number 32
Where Γ (k) is a gain matrix (order m + n + d) that determines the identification / estimation speed of the adaptive parameter, and e asterisk (k) is a signal indicating the identification / estimation error, and is a recurrence as shown in Equations 33 and 34, respectively. It is expressed by an equation.

[0196]

(Equation 32)

[0197]

[Equation 33]

[0198]

(Equation 34)

Various specific algorithms are given depending on how to select λ1 (k) and λ2 (k) in Expression 33.
For example, assuming that λ1 (k) = 1, λ2 (k) = λ (0 <λ <2), a gradually decreasing gain algorithm (least square method when λ = 1), λ1 (k) = λ1 (0 <λ1 <1), λ2 (k)
= Λ2 (0 <λ2 <λ), the variable gain algorithm (weighted least squares method when λ2 = 1), λ1
When (k) / λ2 (k) = σ and λ3 is expressed as in Equation 35, if λ1 (k) = λ3, a fixed trace algorithm is obtained. When λ1 (k) = 1 and λ2 (k) = 0, the fixed gain algorithm is used. In this case, as is apparent from Equation 33, Γ (k) = Γ (k−1), and thus Γ
(k) = fixed value of Γ. For time-varying plants such as fuel injection or air-fuel ratios, any of the progressive gain algorithm, the variable gain algorithm, the fixed gain algorithm, and the fixed trace algorithm are suitable.

[0200]

(Equation 35)

As apparent from the above description, this adaptive controller is a recursive type controller that takes into account the dynamic behavior of the controlled object (internal combustion engine). A controller written in recurrence form to compensate for the behavior. More specifically, since it is of the STR type, it can be defined as an adaptive controller having a recursive adaptive parameter adjustment mechanism at the input of the controller.

Here, the feedback correction coefficient KSTR (k)
Is specifically determined as shown in Expression 36.

[0203]

[Equation 36]

The required fuel injection amount Tcyl is multiplied by the obtained feedback correction coefficient KSTR based on the adaptive control law as the feedback correction coefficient KFB, and the output fuel injection amount Tout (operating amount) is determined and input to the control plant. That is,
As shown in the block diagram of FIG. 8 (and partially shown in the block diagram of FIG. 50), the output fuel injection amount Tout is determined by Tout = Tcyl × KTOTAL × KCMDM × KFB + TTOTAL. The output fuel injection amount Tout has a feedback correction coefficient #nKLAF for each cylinder based on the PID control law.
Is also multiplied, which was previously described with reference to FIG. Further, in the above, TTOTAL indicates the total value of various correction values performed in an addition term such as atmospheric pressure correction. (However, the invalid time of the injector is separately added when the output fuel injection amount Tout is output, and therefore is not included. Not).

The characteristic of FIG. 50 (and FIG. 8) is that
First, the STR controller is placed outside the fuel injection amount calculation system, and the target value is not the fuel injection amount but the air-fuel ratio. That is, the manipulated variable is indicated by the fuel injection amount, and the parameter adjustment mechanism operates so that the detected air-fuel ratio generated in the exhaust system matches the target air-fuel ratio, and the feedback correction coefficient
The point is that the KSTR has been determined and the robustness to disturbance has been improved. However, since this point is described in an application proposed by the present applicant (Japanese Patent Application No. 6-66,594), a detailed description is omitted.

The second feature is that the manipulated variable is determined by multiplying the basic value by the feedback correction coefficient.
Thereby, the convergence of the control is remarkably improved. On the other hand, the configuration has a disadvantage that the control amount is likely to oscillate if the operation amount is not appropriate. The third point of the feature is that a conventional PID controller (referred to as a PID controller) is provided together with the STR controller, a feedback correction coefficient KLAF is determined according to a PID control law, and a final value of the feedback correction coefficient is determined via a switching mechanism. This means that either KSTR or KLAF is selected as KFB.

It should be noted that the PID controller, ie, P
The feedback correction coefficient KLAF according to the ID control law is calculated as follows. First, a control deviation DKAF between the target air-fuel ratio correction coefficient KCMD and the detected air-fuel ratio KACT is obtained as DKAF (k) = KCMD (kd) -KACT (k) (where d is the actually injected fuel detected by the LAF sensor). Is equivalent to the dead time until the application is completed). In this specification, (k) indicates a time (operation or control cycle), and more specifically, indicates a program start time.
(kd): target air-fuel ratio (of control cycle before dead time), KACT
(k): Indicates the detected air-fuel ratio (of the current control cycle).

Next, the P term KL is multiplied by a predetermined coefficient.
AFP (k), I-term KLAFI (k), and D-term KLAFD (k) are converted to P-term: KLAFP (k) = DKAF (k) × KP I-term: KLAFI (k) = KLAFI (k−1) + DKAF ( k) × KI D term: KLAFD (k) = (DKAF (k) −DKAF (k-1) × KD. Thus, the P term is obtained by multiplying the deviation by the proportional gain KP, and the I term is obtained by the deviation. The value obtained by multiplying by the integral gain KI is obtained by adding to the previous value KLAFI (k-1) of the feedback correction coefficient, and the D term is the present value DKAF (k) and the previous value DKAF (k-1) of the deviation.
Multiplied by the differential gain KD. In addition, each gain KP, K
I and KD are obtained in accordance with the engine speed and the engine load. More specifically, the engine speed Ne and the intake pressure P are determined using a map.
b.

Finally, the value thus obtained is added to KLAF (k) = KLAFP (k) + KLAFI (k) + KLAFD (k) to obtain the present value KLAF (k) of the feedback correction coefficient based on the PID control law. In this case, in order to obtain a feedback correction coefficient by multiplication correction, it is assumed that the offset value 1.0 is included in the I term KLAFI (k) (that is, the initial value of the I term KLAFI is 1.0). ). When the feedback correction coefficient is selected by the PID controller, the STR controller holds the adaptive parameter so that the feedback correction coefficient KSTR stops at 1 (initial state).

On the premise of the above, the calculation of the feedback correction coefficient will be described with reference to the flow chart of FIG. The program in FIG. 49 is started at a predetermined crank angle.

First, the engine speed Ne and the intake pressure Pb detected in S700 are read out, and the program proceeds to S704 in which it is determined whether or not fuel cut has been performed. The fuel cut is performed in a predetermined operating state, for example, when the throttle opening is in a fully closed position and the engine speed is equal to or higher than a predetermined value.The fuel supply is stopped, and the air-fuel ratio is controlled in an open loop. Is done.

If it is determined in step S704 that the fuel cut is not performed, the flow advances to step S706 to read the required fuel injection amount Tcyl.
4 is determined. This is for example
When the sensor cell voltage (reference voltage) of the LAF sensor 54 is smaller than a predetermined value (for example, 1.0 V), it is determined that the activation is completed.

If it is determined in step S708 that activation has been completed, the flow advances to step S710 to determine whether or not the current state is in the feedback control area. This is performed in a separate routine that is not disclosed, and is controlled in an open loop, for example, when the engine is fully increased, when the engine speed is high, or when the operating state is suddenly changed due to the influence of EGR or the like.

When the result in S710 is affirmative, the program proceeds to S71.
2, the detected exhaust air-fuel ratio is read, and the routine proceeds to S714, where a detected air-fuel ratio KACT (k) is obtained from the detected exhaust air-fuel ratio. The routine proceeds to S716, where the final value of the feedback correction coefficient is calculated.
Ask for KFB.

FIG. 51 is a subroutine flowchart showing the operation.

Referring to the figure, in S800, it is determined whether or not the open loop control was performed last time (previous control or calculation cycle, that is, last program start time). If the last time was in open loop control such as fuel cut, the result is affirmative and the routine proceeds to S802, where the counter value C is reset to 0, and the routine proceeds to S804 where the flag FKST
The bit of R is reset to 0, and the flow advances to S806 to calculate the final value KFB of the feedback correction coefficient. S80
Resetting the bit of the flag FKSTR to 0 with 4 is
It means that the feedback correction coefficient should be determined by the PID control law. As described later, when the bit of the flag FKSTR is set to 1, it means that the feedback correction coefficient should be determined by the adaptive control law.

FIG. 52 is a subroutine flowchart showing a specific operation of the feedback correction term KFB calculation. In the following, it is determined in S900 whether the bit of the flag FKSTR is set to 1, that is, whether or not the flag is in the STR (controller) operation area. Since this flag has been reset to 0 in S804 of the flow chart of FIG. 51, the determination in this step is denied, and the flow advances to S902 to determine whether the bit of the previous flag FKSTR has been set to 1, that is, the previous STR ( (Controller) It is determined whether or not it is in the operation area.

The determination here is naturally denied, and the process proceeds to S904, where the feedback correction coefficient KLAF (k) is calculated as described above based on the PID control law of the PID controller, more precisely, the feedback correction coefficient calculated by the PID controller. Select the coefficient KLAF (k). Subsequently, returning to the flow chart of FIG. 51, the process proceeds to S808, where KLAF (k) is converted to KF
B.

Continuing the description of the flow chart of FIG. 51, if it is determined in S800 that the control is not the previous open-loop control, that is, that the control has returned from the open-loop control to the feedback control, the flow proceeds to S810 to set the target air-fuel ratio. The difference D between the value KCMD (kd) before the dead time and the current value KCMD (k)
KCMD is obtained and compared with the reference value DKCMDref. And the difference D
When it is determined that KCMD exceeds the reference value DKCMDref,
Proceeding to 802 and thereafter, a feedback correction coefficient is calculated according to the PID control law. This is because when the change in the target air-fuel ratio is large, it is difficult to say that the detected value always indicates a true value due to the detection delay of the air-fuel ratio sensor, as in the case of the return of the fuel cut. This is because it may become unstable. Examples of the case where the change in the target air-fuel ratio is large include, for example, when returning from the full throttle increase, when returning from the lean burn control (for example, when the air-fuel ratio is 20: 1 or leaner) to the stoichiometric air-fuel ratio control, When returning from the perturbation control in which the target air-fuel ratio is amplitude to the stoichiometric air-fuel ratio control in which the target air-fuel ratio is kept constant, there may be mentioned.

On the other hand, in step S810, the difference DKCMD is equal to the reference value DKCMDr.
When it is determined that the temperature is equal to or less than ef, the process proceeds to S812 to increment the counter value C, and proceeds to S814 to detect the detected water temperature Tw.
Is compared with a predetermined value TWSTR.ON, and when it is determined that the value is lower than the predetermined value, the process proceeds to S804 and thereafter to calculate a feedback correction coefficient according to a PID control law. This is because combustion is not stable at a low water temperature, and there is a risk of misfire or the like, and a stable detection value KACT cannot be obtained. When the water temperature is abnormally high, the feedback correction coefficient is calculated by the PID control law for the same reason.

If it is determined in step S814 that the detected water temperature is equal to or higher than the predetermined value, the flow advances to step S816 to compare the detected engine speed Ne with a predetermined value NESTRLMT.
Proceeding to 804 and thereafter, a feedback correction coefficient is calculated according to the PID control law. This is because the calculation time tends to be insufficient at a high rotation speed, and the combustion is not stable.

If it is determined in S816 that the detected engine speed is less than the predetermined value, the flow advances to S818 to determine which valve timing characteristic has been selected, and determines that the HiV / T side characteristic has been selected. When it is determined, the process proceeds to S804 and thereafter to calculate a feedback correction coefficient according to the PID control law. This is because when the characteristic on the HiV / T side is selected, the amount of overlap of the valve timings is large, and there is a possibility that intake air passes through the exhaust valve and escapes, that is, a phenomenon called so-called intake air blow-through occurs. This is because the detected value KACT cannot be expected.

If it is determined in step S818 that the valve is on the LoV / T side (including one of the two valves in the rest state), the process proceeds to step S820.
Then, it is determined whether or not the vehicle is in the idle region. When the result is affirmative, the process proceeds to S804 and thereafter to calculate the feedback correction coefficient according to the PID control law. This is because the operating state is almost stable at the time of idling and does not require a high gain such as the STR control law. In addition, since the intake air amount is controlled using an electric air control valve, so-called EACV, so as to keep the engine speed constant during idling, there is a possibility that the intake air amount control and the air-fuel ratio feedback control may interfere with each other. In this sense, the gain is set relatively low based on the PID control law.

When it is determined in S820 that the detected intake pressure Pb is not in the idle range, the flow proceeds to S822, in which it is determined whether the detected intake pressure Pb is a value on the low load side.
Proceeding to 804 and thereafter, a feedback correction coefficient is calculated according to the PID control law. This is also because the combustion is not stable.

If it is determined in step S822 that the load is not low, the flow advances to step S824 to set the counter value C to a predetermined value, for example, 5
Compare with S804, S806, S900, S902 as long as the counter value C is determined to be equal to or less than the predetermined value.
(S916), proceeding to S904 and S808, the feedback correction coefficient KLAF (k) calculated by the PID controller is selected in the same manner as described above.

That is, in FIG. 48, the time T1 when the fuel cut ends and the control returns from the open loop control to the feedback control (the counter value C = 1 in FIG. 51).
In the period from to T2 (counter value C = 5), the feedback correction coefficient is a value KLAF according to the PID control law determined by the PID controller. This PID
The feedback correction coefficient KLAF based on the control law, unlike the feedback correction coefficient KSTR based on the STR controller, has a characteristic that it does not attempt to absorb the control deviation DKAF between the target value and the detected value at once, but absorbs it relatively slowly.

Therefore, even when the difference between the delay until the completion of the combustion of the resupplied fuel as shown in FIG. 48 and the detection delay of the air-fuel ratio sensor is relatively large, the correction coefficient is S.
The control amount (plant output) does not become unstable as in the case of the TR controller. Here, the predetermined value is 5, in other words, 5
The control cycle is set because it is considered that the combustion delay and the detection delay described above can be absorbed in this period. This period (predetermined value) may be determined from the engine speed, the engine load, etc., which are the exhaust gas transport delay parameters. For example, when the exhaust gas transport delay parameter is small in accordance with the engine speed and the intake pressure, When the predetermined value is small and the exhaust gas transport delay parameter is large, the predetermined value may be set large.

Returning to the description of the flow chart in FIG.
If it is determined in step S824 that the counter value C exceeds the predetermined value, that is, if it is determined that the value is 6 or more, the process proceeds to step S826, and the flag FKST
The bit of R is set to 1, and the flow advances to S828 to return to FIG.
Calculate the final value KFB of the feedback correction coefficient according to the two flow charts. In this case, the determination in S900 is affirmed in the flow chart of FIG. 52, and the flow advances to S906 to determine whether the bit of the previous flag FKSTR has been reset to 0, that is, whether or not the last time was in the PID operation area.

When the counter value has exceeded the predetermined value for the first time, this determination is affirmative, and the routine proceeds to S908, where the detected air-fuel ratio is determined.
KACT (k) is compared with a lower limit value a, for example, 0.8. When it is determined that the detected air-fuel ratio is equal to or more than the lower limit value, the process proceeds to S910, and the detected air-fuel ratio is compared with the upper limit value b, for example, 1.2. When it is determined that the detected air-fuel ratio is lower than S912, the process proceeds to S914 via S912. Is used to calculate the feedback correction coefficient KSTR (k). More precisely, the feedback correction coefficient KSTR (k) calculated by the STR controller is selected.

In other words, if it is determined in S908 that the detected air-fuel ratio is lower than the lower limit value a, or if it is determined in S910 that the detected air-fuel ratio is higher than the upper limit value b, the process proceeds to S904 and the PI
A feedback correction coefficient is calculated based on the D control.
That is, the switching from the PID control to the STR (adaptive) control is performed in the operation range of the STR controller and the detected air-fuel ratio.
KACT is performed when it is near 1. As a result, switching from PID control to STR (adaptive) control can be smoothly performed, and oscillation of the control amount can be prevented.

If it is determined in step S910 that the detected air-fuel ratio KACT (k) is equal to or smaller than the upper limit b, the process proceeds to step S912, in which ST
In the R controller, the scalar amount b ₀ for determining the gain is divided by the previous value KLAF (k−1) of the feedback correction coefficient by the PID control as shown in FIG.
Proceeding to step 4, the STR controller obtains a feedback correction coefficient KSTR (k).

That is, the feedback correction coefficient KSTR (k) by the STR controller is originally obtained as shown in Expression 36 as described above.
When proceeding thereafter, the feedback correction coefficient is determined based on the PID control in the previous control cycle. And
In the configuration of FIG. 50, when the feedback correction coefficient is determined by the PID control, the STR controller stops the operation by setting the feedback correction coefficient KSTR to 1, as described above. In other words, the adaptive parameter (vector) θ hat (k) used in the STR controller is KSTR
= 1.0. Therefore, when the STR controller determines the feedback correction coefficient KSTR again, if the value of KSTR greatly deviates from 1, the control amount becomes unstable. Therefore, the scalar quantity b ₀ that determines the gain in the adaptive parameter θ hat (k) held so that KSTR becomes 1.0 (initial value) or near 1.0 is set to the feedback correction coefficient of the PID control. By dividing by the previous value, for example, if the combination of adaptive parameters is set to KSTR = 1.0, Equation 37
Since the first term is 1, as shown in FIG.
The value of the term KLAF (k-1) becomes the current correction coefficient KSTR (k). As a result, in addition to setting the detection value KACT to 1 or a value close to it in S908 and S910, the PID control also causes
Switching to R control can be performed more smoothly.

[0233]

(37)

To supplement the explanation of the flow chart of FIG. 52, if it is determined in S902 that the STR (controller) is in the previous operation area, the flow advances to S916 to execute the previous value KSTR (k-1) of the feedback correction coefficient by the STR controller.
Is the previous value KLAFI (k-1) of the I term. As a result, S90
When KLAF (k) is calculated in 4, the I term, KLAFI
Is KLAFI (k) = KSTR (k-1) + DKAF (k) × KI, and the obtained I term is added to the P and D terms to obtain KLAF (k).

That is, when the feedback control coefficient is calculated by switching from the adaptive control to the PID control, the integral term may change abruptly. Thus, the KID value is used to calculate the PID control correction coefficient. By determining the initial value, the difference between the correction coefficient KSTR (k-1) and the correction coefficient KLAF (k) can be kept small, so that when switching from the STR control to the PID control, the value of the feedback correction coefficient can be reduced. , The difference between the control amounts can be made small, and the continuous control can be performed smoothly, and a sudden change in the control amount can be prevented.

In the flow chart of FIG. 52, S
At 900 the STR (controller) operation area is determined,
If it is also determined in S906 that the current time is not the PID operation area, the process proceeds to S914, where the feedback correction coefficient KSTR (k) is calculated based on the STR controller. As mentioned.

Returning to the flow chart of FIG. 51, the program proceeds to S830, in which it is confirmed whether or not the correction coefficient obtained in the flow chart of FIG. 52 is KSTR. Difference from 0 (1-KSTR (k))
And calculate its absolute value to a predetermined threshold value KSTR
Compare with ref.

That is, when the feedback correction coefficient fluctuates greatly, the control amount also changes suddenly, and the stability of the control decreases. Therefore, the absolute value of the difference between the obtained feedback correction coefficient and 1.0 is compared with the threshold value, and if it exceeds the threshold value, the process proceeds to S804, and the feedback correction coefficient is determined again based on the PID control. Thus, stable control can be realized without a sudden change in the control amount. In this case, the comparison was made using the absolute value of the difference between the feedback correction coefficient and 1.0. However, as shown in FIG. May be. If the absolute value of the feedback correction coefficient KSTR (k) obtained in S832 does not exceed the threshold value, the process proceeds to S834, and the value by the STR controller is used as the feedback correction coefficient KFB. Also,
When the result in S830 is NO, the program proceeds to S836, in which the flag FK is set.
The bit of STR is reset to 0, and the process proceeds to S838, where PI
The value obtained by the D controller is used as the final value KFB of the feedback correction coefficient.

Returning to the flow chart of FIG. 49, the program then proceeds to S718 in which the required fuel injection amount Tcyl is multiplied by the final value KFB or the like of the feedback correction coefficient obtained.
The output fuel injection amount Tout is determined by adding the added value TTOTAL. Next, the process proceeds to S720 to perform the intake pipe wall adhesion correction (described later), and proceeds to S722 to output the output fuel injection amount Tout (n) to the injector 22 as an operation amount. Here, n means a cylinder, and the output fuel injection amount Tout is finally determined for each cylinder.

If it is determined in step S704 that the fuel cut has been made, the flow advances to step S728 to set the output fuel injection amount Tout to zero. Since the air-fuel ratio is open-loop control when a negative in S710 to no S 708, S72 6
Then, the final value KFB of the feedback correction coefficient is set to 1.0, and the routine proceeds to S718, where the output fuel injection amount Tout is obtained. When the result in S704 is affirmative, open-loop control is also performed, and the output fuel injection amount Tout is set to a predetermined value (S7).
28).

In the above, when the open-loop control of the air-fuel ratio such as when returning from fuel cut is ended and the feedback control is restarted, the PI is maintained for a predetermined period.
Since the feedback correction coefficient is determined based on the D control law, it takes time for the supplied fuel to burn, or because the sensor itself has a detection delay, the detected air-fuel ratio and the actual When there is a relatively large difference between the air-fuel ratio and the air-fuel ratio, the control amount (air-fuel ratio) can be destabilized without using the feedback correction coefficient by the STR controller, thereby reducing the control stability. Absent.

On the other hand, since the period is set to a predetermined value, when the detected value is stabilized, the operation is performed so as to absorb at once the control deviation between the target air-fuel ratio and the detected air-fuel ratio using the feedback correction coefficient by the STR controller. , The convergence of the control can be improved. In particular, in the embodiment, the convergence of the control is improved so that the manipulated variable is determined by multiplying the basic value by the feedback correction coefficient. Therefore, the stability and the convergence of the control are more preferably balanced. Can be done. Note that the air-fuel ratio detected immediately after the activation of the LAF sensor 54 is not stable.
The feedback correction coefficient may be determined based on the PID control rule for a predetermined period after the activation of the sensor 54.

Further, when the fluctuation of the target air-fuel ratio is large,
Since the feedback correction coefficient is determined based on the PID control even after a predetermined period has elapsed, the stability and convergence of the control are not limited to fuel cut, but also when returning from open loop control such as full-open increase. Can be optimally balanced. Further, when the feedback correction coefficient by the STR controller becomes unstable, the feedback correction coefficient is determined based on the PID control law, so that the control stability and the convergence can be more optimally balanced. .

In particular, at the time of shifting from the STR control to the PID control, at least a part of the element, that is, the I term is calculated by using the feedback correction coefficient by the STR controller. It is possible to effectively prevent the control amount from oscillating due to a sudden change in the operation amount due to a step in the correction coefficient. Therefore, it is possible to effectively prevent the control stability from being lowered.

Further, when returning from the PID control to the STR control, the case where the detected value KACT is at or near 1 is selected, and the first value of the feedback correction coefficient by the adaptive control law (STR controller) is determined by the PID control law. Since the feedback correction coefficient is substantially the same as the feedback correction coefficient, when switching from PID control to STR control, the switching can be performed smoothly. Thereby,
It is possible to effectively prevent the control amount from becoming unstable due to a sudden change in the manipulated variable due to the occurrence of a step in the correction coefficient, and it is possible to effectively prevent the control stability from deteriorating.

Here, the correction of the adhesion of the output fuel injection amount Tout to the wall surface of the intake pipe will be described. As described above, the intake pipe wall surface adhesion correction is performed for each cylinder, and the cylinder number n (n = 1, 2, 2)
(3) and (4) are specified.

In order to respond to the change of the adhesion parameter,
Before the wall deposition plant, a wall deposition correction compensator having an inverse transfer function is inserted in series. The adhesion parameter of the wall adhesion correction compensator is searched by a map determined in advance based on the correspondence with the engine operating state.

If the adhesion parameter of the wall adhesion correction compensator is equal to the true adhesion parameter of the actual machine,
Both have a transfer function of 1 when viewed from the outside, that is, the product of the transfer function of the plant and the compensator is 1, and the target cylinder intake fuel amount = the cylinder actual intake fuel amount, so that a complete correction should be made. .

Assuming the above, the subroutine flow chart shown in FIG. 54 for the work of correcting the wall adhesion of the output fuel injection amount Tout in S720 of the flow chart of FIG.
This will be described with reference to a chart. This routine is executed by TDC
The output fuel injection amount Tou for all cylinders is performed in synchronization with the signal.
Until t (n) is obtained, the processing is performed for the number of cylinders. Also,
Although (k-1) in the figure indicates the previous calculated value for the cylinder n as described above, the description of (k) is omitted for the current calculated value.

First, various parameters are read in S1000, and the process proceeds to S1002, where the direct rate A and the carry-out rate B
Ask for. This is performed by searching a map showing the characteristics in FIG. 55 from the engine speed Ne and the intake pressure Pb. Note that this map is set separately according to the valve timing characteristics of the variable valve timing mechanism, and a map corresponding to the currently selected valve timing characteristics is searched for. At the same time, a table showing the characteristics shown in FIG. 56 is searched from the detected water temperature Tw to obtain the correction coefficients KATW and KBTW, and the correction is performed by multiplying the map search value. Although not shown, other similar correction coefficients KA and KB are obtained according to the presence or absence of execution of EGR or canister purge and the magnitude of the target air-fuel ratio KCMD. Specifically, it is as follows. Ae = A × KATW × KA Be = B × KBTW × KB The corrected direct rate A is Ae, and the carry-out rate B is Be.

Subsequently, the flow proceeds to S1004 to judge whether or not fuel cut is performed. If the result is NO, the flow proceeds to S1006 to correct the output fuel injection amount Tout as shown in FIG. F is obtained, and if affirmative, the routine proceeds to S1008, where the output fuel injection amount Tout for each cylinder
(n) -F is set to zero. Here, the value TWP (n) is the amount of fuel attached to the intake pipe.

FIG. 57 is a flowchart for calculating the intake pipe adhering fuel amount TWP (n), which is started at a predetermined crank angle.

First, in S1100, the current program activation is a period from the start of calculation of the fuel injection amount Tout to the end of fuel injection of any one of the cylinders (hereinafter referred to as “injection control period”).
Is determined, and if affirmative, the routine proceeds to S1102, where the bit of the first flag FCTWP (n) indicating the end of the calculation of the amount of adhering fuel of the cylinder is set to 0, and the amount of adhering fuel is set. Allow the operation and end the program. When the result in S1100 is NO, the program proceeds to S1104, in which the first flag FCTW is set.
It is determined whether or not the bit of P (n) is 1. If the result is affirmative, the calculation of the amount of fuel adhering to the cylinder has already been completed. Therefore, the process proceeds to S1106.
The routine proceeds to 108, where it is determined whether or not the fuel is cut.

When the result in S1108 is NO, S1110
Then, the intake pipe adhering fuel amount TWP (n) is calculated as shown in the figure. Here, TWP (k-1) is the previous value of TWP (k). The first term on the right-hand side means the amount of fuel remaining before being removed this time out of the fuel previously deposited, and the second term on the right-hand side.
The term means the amount of fuel newly attached to the intake pipe among the fuel injected this time. Then, the process proceeds to S1112, where the bit of the second flag FTWPR (n) indicating that the amount of deposited fuel is zero is set to 0, and the process proceeds to S1106, where the first flag FC
Set the TWP (n) bit to 1 and end the program.

If it is determined in S1108 that the fuel is cut, the flow advances to S1114 to determine whether or not the bit of the second flag FTWPR (n) indicating that the remaining amount of deposited fuel is zero is 1, and affirmatively. Since the amount of fuel adhering is zero (TWP (n) = 0), the process proceeds to S1106,
If the result in step S1116 is negative, the program proceeds to step S1116, where the amount of deposited fuel TWP (n) is calculated from the equation shown. Here, the equation shown is S1
This corresponds to the expression 110 with the second term on the right side deleted.
This is because the fuel is being cut and no new fuel is attached.

Subsequently, the flow proceeds to S1118, in which it is determined whether the TWP (n) value is larger than the minute predetermined value TWPLG. If the result is affirmative, the process proceeds to S1112. Since the power is less, the process advances to S1120 to set TWP (n) = 0, the process advances to S1122, sets the bit of the second flag FTWPR (n) to 1, and advances to S1106.

In this way, the fuel amount TWP (n) attached to the intake pipe for each cylinder can be calculated with high accuracy.
By using the TWP (n) value in the calculation of the fuel injection amount Tout in FIG. 54, an appropriate amount of fuel is taken into consideration for each cylinder in consideration of the amount of fuel attached to the intake pipe and the amount of fuel removed from the attached fuel. Can be supplied to the room. In the above, even in the engine start mode (including simultaneous injection and sequential injection), the direct rate A and the carry-out rate B
Then, the calculation of the amount of fuel TWP attached to the intake pipe is started, and the attachment correction is executed.

In this embodiment, as described above, the fuel injection amount control means for controlling the fuel injection amount of the internal combustion engine and the exhaust system of the internal combustion engine are arranged upstream of the catalyst device (28). First to detect the air-fuel ratio of exhaust gas to be exhausted
And the air-fuel ratio KACT detected by the first air-fuel ratio detecting means (LAF sensor 54) and the target air-fuel ratio KCMD.
Fuel injection correction amount calculating means for calculating a fuel injection correction amount so as to coincide with the following, and second air-fuel ratio detecting means disposed downstream of the catalyst device and detecting an air-fuel ratio of exhaust gas passing through the catalyst. (O ₂ sensor 56), the fuel injection correction amount calculating means sets the air-fuel ratio detected by the first air-fuel ratio detecting means to coincide with the target air-fuel ratio. An adaptive controller (STR controller) for calculating a fuel injection correction amount, an adaptive parameter adjusting mechanism for adjusting an adaptive parameter input to the adaptive controller, and an air-fuel ratio detected by the second air-fuel ratio detecting means. Correction means for correcting the target air-fuel ratio KCMD, so that it is possible to adaptively compensate for the dynamic behavior of the air-fuel ratio caused by the aging of the internal combustion engine and the solid-state variation. Come, the target value is determined based on the air-fuel ratio detected by the second air-fuel ratio detection means, it is possible to match the air-fuel ratio immediately.

That is, in this embodiment, the adaptive control for calculating the correction coefficient KSTR of the fuel injection amount so that the air-fuel ratio detected by the first air-fuel ratio detecting means (LAF sensor 54) matches the target air-fuel ratio KCMD. (STR controller) and STR
Together and an adaptive parameter-adjusting mechanism for adjusting the adaptive parameters to be input to the controller, correction means for correcting the target air-fuel ratio KCMD according to the air-fuel ratio detected by the second air-fuel ratio detecting means (O ₂ sensor 56) Because it was configured to be equipped, the behavior of the air-fuel ratio can be dynamically compensated,
The air-fuel ratio can be instantaneously matched with the target air-fuel ratio.

In FIG. 8, a third catalyst device 94 may be arranged in a block 400 indicated by an imaginary line upstream of the LAF sensor 54. The third catalyst device 94 is preferably a so-called light-off catalyzer (early active catalyzer). Also, the third catalyst device 9
4 may have a sufficiently smaller capacity than the downstream catalytic device. Furthermore, a three-way catalyst type similar to the downstream catalyst device may be used, or an electrically heated catalyst called EHC (Electric Heated Catalyzer) may be activated early. The third catalyst device 94 may be provided as needed. Particularly, when the above-described system is configured for each bank of the V-type engine, the exhaust volume is relatively reduced. This is effective when the temperature rise is slow. Incidentally, when the third catalyst device 94 is arranged, since the dead time and the like differ, it goes without saying that the control amount and the like differ.

In FIG. 8, a filter 96 may be arranged in front of the observer as shown by an imaginary line. Since there is a response delay in the LAF sensor 54, the observer deals with the internal calculation as described above, but as shown in FIG.
A filter that compensates for the next delay characteristic (ie, a leading filter)
96 may be arranged to deal with hardware.

It should be noted that the configuration shown in the block diagram of FIG. 8 is not all essential to the present invention. For example, the basic fuel injection amount may be obtained by a method other than the disclosed method, and the adhesion correction is not essential for the present invention.

FIG. 58 is a block diagram similar to FIG. 8, showing a second embodiment of the apparatus according to this application.

In the second embodiment, as shown in FIG.
A second O ₂ sensor 98 was arranged downstream of the second catalyst device 30. The detection output of the second O ₂ sensor 98 is used for correcting the target air-fuel ratio KCMD as shown in the figure. Thereby, the target air-fuel ratio KCMD can be further optimally set,
Controllability is improved. Further, by detecting the air-fuel ratio in the exhaust gas finally discharged to the atmosphere, the emission performance is improved, and the deterioration state of the catalyst device upstream of the second O ₂ sensor can be monitored. . The second O
_{The second} sensor 98 may be used in place of the first O ₂ sensor 56. Further, the second O ₂ sensor 98 may be mounted in a multi-stage second catalyst device as shown in FIG. 5, similarly to the first O ₂ sensor 56.

In this case, a low-pass filter 500 having a frequency characteristic of about 1000 Hz is connected to the next stage of the second O ₂ sensor 98. The filter 60 of the first O ₂ sensor 56 and the filter 50 of the second O ₂ sensor 98
For 0, a filter such as a linearizer may be used to compensate for the non-linear characteristic.

In the above-described first and second embodiments, the mechanism for driving the throttle valve 16 via the pulse motor M has been described. However, similar to a generally known mechanism,
It may be mechanically linked to the accelerator pedal.

Although a responsive motorized exhaust gas recirculation valve is used for the exhaust gas recirculation mechanism, an exhaust gas recirculation valve using a diaphragm operated by the negative pressure of the engine may be used.

The second catalyst device 30 may not be provided depending on the purification performance of the first catalyst device 28.

Further, although a low-pass filter is used, a band-pass filter which can obtain equivalent performance may be used.

Further, in the above-described configuration, an example has been shown in which the air-fuel ratio of each cylinder is estimated using one air-fuel ratio sensor and is controlled to the target value. However, the present invention is not limited to this. A fuel ratio sensor may be provided to directly detect the air-fuel ratio of each cylinder.

In the above embodiment, the air-fuel ratio is actually obtained as an equivalent ratio, but this is exactly the same as using the air-fuel ratio itself.

In the above description, the feedback correction coefficients KSTR to KLAF are obtained as multiplication values, but may be obtained as addition values.

In the above, ST is used as the adaptive controller.
Although R has been described as an example, MRACS (model reference adaptive control) may be used.

[0274]

According to the fuel injection control device for an internal combustion engine of the first aspect, the air-fuel ratio can be dynamically compensated for the behavior, and is determined based on the air-fuel ratio detected by the second air-fuel ratio detecting means. The air-fuel ratio can be instantaneously matched with the target value to be set.

[0275] According to the present invention, as compared with the case where the device is disposed downstream of the catalyst device, the time during which the output is inverted is shortened, and the detection accuracy and, consequently, the control accuracy are improved.

In the present invention, noise can be removed by appropriately selecting the frequency characteristics of the filter, and the detection accuracy is improved and the controllability is improved.

According to the present invention, the response time can be optimized by appropriately selecting the frequency characteristics of the filter, and the detection accuracy is improved and the controllability is improved.

According to the fifth aspect, the frequency characteristics of the filter can be optimized to reliably remove noise, or the response time can be optimized, so that the detection accuracy is improved and the controllability is improved. improves.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a fuel injection control device for an internal combustion engine according to the present application as a whole.

FIG. 2 is an explanatory diagram showing details of an exhaust gas recirculation mechanism in FIG. 1;

FIG. 3 is an explanatory diagram showing details of a canister / purge mechanism in FIG. 1;

FIG. 4 is an explanatory diagram showing valve timing characteristics of the variable valve timing mechanism in FIG.

FIG. 5 is an explanatory diagram showing an arrangement configuration of a first catalyst device and an O ₂ sensor in FIG. 1;

FIG. 6 is a block diagram showing details of a control unit in FIG. 1;

FIG. 7 is an explanatory diagram showing an output of an O ₂ sensor in FIG. 1;

FIG. 8 is a functional block diagram showing an operation of a fuel injection control device for an internal combustion engine according to the present application.

FIG. 9 is a flowchart showing a calculation operation of a basic fuel injection amount TiM-F in the block diagram of FIG. 8;

10 is a basic fuel injection amount TiM in the flow chart of FIG. 9;
It is a block diagram explaining the calculation operation of -F.

FIG. 11 is a block diagram showing a method for calculating an effective opening area of a throttle valve using a flow coefficient or the like.

FIG. 12 is an explanatory diagram showing map characteristics of coefficients used in the calculation in FIG. 11;

13 is an explanatory diagram showing map characteristics of a fuel injection amount Timap in a steady operation state used in the flow chart of FIG. 9 and the block diagram of FIG. 10;

FIG. 14 is an explanatory diagram showing a map characteristic of a target air-fuel ratio used in the flow chart of FIG. 9 and the block diagram of FIG. 10, more specifically, a basic value thereof.

FIG. 15 is a data diagram showing a simulation result of an effective opening area of a throttle in a calculation operation of the basic fuel injection amount TiM-F in the flow chart of FIG. 9 and the block diagram of FIG.

16 is an explanatory diagram showing a steady operation state and a transient operation state in the calculation operation of the basic fuel injection amount TiM-F in the flow chart of FIG. 9 and the block diagram of FIG.

FIG. 17 is an explanatory diagram showing the relationship between the throttle opening and the effective opening area of the throttle in the calculation of the basic fuel injection amount TiM-F in the flow charts of FIGS. 9 and 10;

FIG. 18 is a basic fuel injection amount TiM in the flow chart of FIG. 9;
It is a block diagram explaining the example of correction of the calculation operation of -F.

FIG. 19 is a flowchart showing an operation of estimating an exhaust gas recirculation rate in calculating an EGR correction coefficient in the block diagram of FIG. 8;

20 is an explanatory diagram showing a basic algorithm for estimating the exhaust gas recirculation rate, and is an explanatory diagram showing characteristics of a gas amount with respect to a lift amount of the exhaust gas recirculation rate used in the calculation of the flow chart of FIG. 19;

FIG. 21 is an explanatory diagram showing delay of an actual lift and recirculation gas with respect to a lift command value of an exhaust gas recirculation valve.

FIG. 22 is an explanatory diagram showing map characteristics of an exhaust gas recirculation rate correction coefficient (basic exhaust gas recirculation rate correction coefficient) in a steady state used for the calculation of the flow chart of FIG. 19;

FIG. 23 is an explanatory diagram showing map characteristics of a lift command value used for calculation of the flow chart of FIG. 19;

FIG. 24 is a subroutine flowchart showing a calculation operation of a fuel injection correction coefficient in the flowchart of FIG. 19;

FIG. 25 is an explanatory diagram showing a configuration of a ring buffer used in the operation of the flow chart of FIG. 24;

FIG. 26 is an explanatory diagram showing a map characteristic of a dead time τ used in the operation of the flow chart of FIG. 24;

FIG. 27 is a timing chart for explaining the operation of the flow chart of FIG. 24;

FIG. 28 is a flowchart showing a calculation operation of a canister purge correction coefficient in the block diagram of FIG. 8;

FIG. 29 is a flowchart showing a calculation operation of a target air-fuel ratio and an air-fuel ratio correction coefficient in the block diagram of FIG. 8;

FIG. 30 is a flowchart showing a correction coefficient KE in the flowchart of FIG. 29;
FIG. 3 is an explanatory diagram showing characteristics of TC.

FIG. 31 is an explanatory diagram showing a relationship between TDC of a multi-cylinder internal combustion engine and an air-fuel ratio of an exhaust system assembly.

FIG. 32 is an explanatory diagram showing quality of sample timing with respect to an actual air-fuel ratio.

FIG. 33 is a flowchart showing a sampling operation of a detected air-fuel ratio in a Sel-V block in the block diagram of FIG. 8;

FIG. 34 is one of explanatory diagrams of the observer in the block diagram of FIG. 8, and is a block diagram showing an example in which the detection operation of the LAF sensor described in the earlier application is modeled.

FIG. 35 is a model obtained by discretizing the model shown in FIG. 34 with a period ΔT.

FIG. 36 is a block diagram showing a true air-fuel ratio estimator modeling the detection behavior of the air-fuel ratio sensor.

FIG. 37 is a block diagram showing a model showing the behavior of the exhaust system of the internal combustion engine.

FIG. 38 is a data diagram showing a case in which fuel is supplied with an air-fuel ratio of three cylinders to 14.7 and an air-fuel ratio of one cylinder to 12.0 for a four-cylinder internal combustion engine using the model shown in FIG. 37; .

39 is a data diagram showing the air-fuel ratio of the aggregate of the model of FIG. 37 when the input shown in FIG. 38 is given.

40 compares data representing the air-fuel ratio of the collective portion of the model of FIG. 37 when the input shown in FIG. 38 is given in consideration of the response delay of the LAF sensor, and the measured value of the LAF sensor output in the same case. It is a data diagram.

FIG. 41 is a block diagram showing a configuration of a general observer.

FIG. 42 is a block diagram showing a configuration of the observer used in the earlier application in the observer shown in the block diagram of FIG. 8;

43 is an explanatory block diagram showing a configuration obtained by combining the model shown in FIG. 37 and the observer shown in FIG. 42;

FIG. 44 is a block diagram showing feedback control of the air-fuel ratio in the block diagram of FIG. 8;

FIG. 45 is an explanatory diagram showing characteristics of a timing map used in the flow chart of FIG. 33;

46 is an explanatory diagram illustrating sensor output characteristics with respect to the engine speed and the engine load, for explaining the characteristics of FIG. 45.

FIG. 47 is a timing chart illustrating a sampling operation in the flow chart of FIG. 33;

FIG. 48 is a timing chart showing a delay in detection of an air-fuel ratio when fuel supply is restarted from fuel cut.

FIG. 49 is a flowchart showing the operation of calculating the feedback correction coefficient in the block diagram of FIG. 8;

FIG. 50 is a block diagram functionally showing the operation of the flow chart of FIG. 49.

FIG. 51 is a subroutine flowchart showing a more specific operation of calculating the feedback correction coefficient in the flowchart of FIG. 49;

FIG. 52 is a similar subroutine flowchart showing a more specific calculation operation of the feedback correction coefficient in the flowchart of FIG. 51.

FIG. 53 is a timing chart illustrating a part of the operation of the flow chart of FIG. 51.

FIG. 54 is a subroutine flowchart for correcting the adhesion of the output fuel injection amount to the intake pipe wall surface in the flow chart of FIG. 49;

FIG. 55 is an explanatory diagram showing map characteristics such as a direct ratio used in the calculation of the flow chart of FIG. 54;

FIG. 56 is an explanatory diagram showing table characteristics of correction coefficients used in the calculation of the flow chart of FIG. 54;

FIG. 57 is a subroutine flowchart showing the operation of calculating TWP (n) in the flowchart of FIG. 54;

FIG. 58 is a block diagram showing a configuration of another embodiment of a fuel injection control device for an internal combustion engine according to the present application.

[Explanation of symbols]

10 internal combustion engine 12 intake pipe 20 intake manifold 22 injector 24 exhaust manifold 26 exhaust pipe 28 first catalyst device 30 and the second catalytic converter 34 control unit 54 wide-range air-fuel ratio sensor (LAF sensor) 56 O ₂ sensor 92 filter 93 filter 94 Third catalyst device 96 Filter 98 Second O ₂ sensor 100 Exhaust gas recirculation mechanism 200 Canister / purge mechanism 300 Variable valve timing mechanism 500 Filter

────────────────────────────────────────────────── ─── Continuing on the front page (51) Int.Cl. ⁷ Identification symbol FI G05B 13/02 G05B 13/02 Z (72) Inventor Isao Komoriya 1-4-1, Chuo, Wako, Saitama, Japan Honda R & D Co., Ltd. In-house (72) Inventor Koichi Nishimura 1-4-1 Chuo, Wako-shi, Saitama Prefecture Inside Honda R & D Co., Ltd. (72) Inventor Toshiaki Hirota 1-4-1 Chuo Wako-shi, Saitama Co., Ltd. (56) References JP-A-6-129283 (JP, A) JP-A-5-26081 (JP, A) JP-A-6-129287 (JP, A) JP-A-6-50204 (JP, A) Kaihei 4-187843 (JP, A) Special Table 3-500565 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) F02D 41/14 F02D 45/00 G05B 11/36 G05B 13/02

Claims (5)

(57) [Claims] 1. A method according to claim 1, Fuel injection amount control means for controlling the fuel injection amount of the internal combustion engine; b. Calculate the fuel injection amount according to the operating state of the internal combustion engine
Means for calculating the amount of fuel injection to be performed ; c . First air-fuel ratio detection means disposed upstream of a catalyst device disposed in an exhaust system of the internal combustion engine, for detecting an air-fuel ratio of exhaust gas discharged from the internal combustion engine; d . As the air-fuel ratio detected by the first air-fuel ratio detection means is equal to the target air-fuel ratio, calculated for the fuel injection amount calculating means
Fuel injection correction factor for correcting the fuel injection correction amount to be out
And the fuel injection correction coefficient calculating means for calculating the number, and e. A second air-fuel ratio detection unit disposed downstream of the catalyst device and detecting an air-fuel ratio of exhaust gas passing through the catalyst device , wherein the fuel injection correction coefficient calculation unit includes: Is f . According to the air-fuel ratio detected by the second air-fuel ratio detecting means
A target air-fuel ratio correcting means for correcting the target air-fuel ratio by means of g . Air-fuel ratio is said to detect the first air-fuel ratio detection means
To match the corrected target air-fuel ratio, the fuel injection correction
An adaptive controller for calculating coefficients ; and h . The adaptive parameter input to the adaptive controller ,
The detected air-fuel ratio detected by the first air-fuel ratio detecting means and the adaptation
A fuel injection control device for an internal combustion engine, comprising: an adaptive parameter adjustment mechanism that adjusts based on past values of parameters. 2. The internal combustion engine according to claim 1, wherein the catalyst device has a multi-stage catalyst bed, and the second air-fuel ratio detecting means is disposed between the multi-stage catalyst beds. Engine fuel injection control device. 3. The fuel injection control device for an internal combustion engine according to claim 1, wherein a filter means is connected to said first air-fuel ratio detecting means. 4. The fuel injection control device for an internal combustion engine according to claim 1, wherein a filter means is connected to the second air-fuel ratio detecting means. 5. The fuel injection control device for an internal combustion engine according to claim 4, wherein said filter means is a low-pass filter.