JPH08312433A - Control device of intube injection type spark ignition system internal combustion engine - Google Patents

Abstract

PURPOSE: To provide the control device of an intube injection type spark ignition system internal combustion engine in which it is not necessary to reset a setting map to set the parameter values giving an influence to the combustion condition in a tube such as the fuel injection amount and the ignition timing at every time of design change of an engine, by consuming the time and the labor. CONSTITUTION: An objective mean effective pressure data is calculated (70c and 70r) depending on the detecting results of operating condition detecting means 17, 29, and 31, to detect at least the engine speed and the load data, parameter values to give the influence to the combustion condition in a combustion chamber are set (70j, 70m, 70n, and 70p) according to the calculated object mean effective pressure data, and the internal combustion engine is controlled depending on the set parameter values.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device for controlling the output of a cylinder injection type spark ignition internal combustion engine mounted on an automobile or the like.

[0002]

2. Description of the Related Art In recent years, as a method for supplying fuel to an internal combustion engine in order to reduce harmful gas components emitted from a fuel injection spark ignition type internal combustion engine mounted on an automobile or to improve fuel efficiency of the engine, , An engine that adopts a cylinder injection type that directly injects fuel into the combustion chamber instead of the conventional intake pipe injection type (hereinafter, cylinder injection gasoline engine)
Have been proposed.

[0003] In a cylinder injection gasoline engine, a lean fuel mixture (lean) is provided as a whole by locally supplying a layered mixture having an air-fuel ratio close to the stoichiometric air-fuel ratio to the periphery of a spark plug or a cavity provided in a piston. It has the advantages that ignition is possible even at various air-fuel ratios, CO and HC emissions are reduced, and fuel efficiency during idle operation and steady running can be greatly improved. Further, the in-cylinder injection gasoline engine has an advantage that the acceleration / deceleration response is also very good because there is no transfer delay due to the intake pipe when the fuel injection amount is increased or decreased. However, at high load, the fuel injection amount increases and the vicinity of the spark plug becomes fuel rich (rich), and the overall air-fuel ratio (also called the average air-fuel ratio)
Has a drawback that a so-called rich misfire occurs and the stable operation region is narrow. This is because it is difficult to change the injection amount and the injection direction of the fuel injection valve per unit time, so it is difficult to maintain the local air-fuel ratio near the spark plug at the optimum value over the entire engine operating range. And so on.

In order to solve such a drawback, fuel injection is performed at an appropriate timing according to the load, and the shape of the combustion chamber is designed according to this, for example,
It is proposed in JP-A-5-79370. More specifically, there has been proposed a method of switching between a late injection mode in which fuel is injected during the compression stroke and an early injection mode in which fuel is injected during the intake stroke, depending on the load. This conventional proposed in-cylinder injection gasoline engine injects fuel into the cavity at the end of the compression stroke and the beginning of the intake stroke during low and medium load operation, and locally approaches the theoretical air-fuel ratio around the spark plug and in the cavity. Form an air-fuel ratio mixture. As a result, ignition is possible even with a lean air-fuel ratio as a whole, CO and HC emissions are reduced, and fuel efficiency during idle operation and steady running is significantly improved. Further, during high load operation, fuel is injected outside the cavity during the intake stroke to form an air-fuel mixture with a uniform air-fuel ratio in the combustion chamber. This makes it possible to burn an amount of fuel equivalent to that of the intake pipe injection type, and to secure the output required at the time of starting and accelerating.

[0005]

When the in-cylinder injection gasoline engine proposed above is used, by switching between the late injection mode and the early injection mode, the overall air-fuel ratio in the late injection mode is a very large value, for example, 25. ~ 40
It is also possible to set to, by supplying a large amount of fresh air from a passage that bypasses the throttle valve or recirculating a large amount of exhaust gas (hereinafter referred to as EGR),
It enables lean combustion during low-load operation such as idling, and can reduce emissions of harmful exhaust gas components and improve fuel efficiency. Further, even if the fresh air intake air amount and the EGR amount are constant, the overall air-fuel ratio can be set to an appropriate value by adjusting the fuel injection amount within the range where misfire does not occur. Even if the amount is the same, the overall air-fuel ratio can be set to an appropriate value even if the fresh air intake air amount or the EGR amount is changed, and it is possible to perform output adjustment regardless of changes in the intake air amount. Become.

In the intake pipe injection type engine, the air-fuel ratio may be operated at a lean air-fuel mixture leaner than the stoichiometric air-fuel ratio. However, even in such lean-burn operation, the air-fuel ratio is at most 20-. Since it is about 24, and the required engine output and the intake air amount supplied to the engine are maintained in a substantially correlative relationship, the intake air amount, such as the pressure in the intake passage downstream of the throttle valve and the air obtained by the air flow sensor, are related. The fuel supply could be determined from the flow rate.

On the other hand, in the output control of the in-cylinder injection type engine in the above-mentioned late injection mode, the air-fuel ratio is 20 to 40.
In order to burn the air-fuel mixture that is extremely lean, there is a part similar to that of the diesel engine, and there is a problem that the load must be determined independently of the intake air amount. For example, when setting the required fuel injection amount, even if the intake air amount measured by the air flow sensor or the pressure in the intake passage downstream of the throttle valve is used as the parameter value, those control parameters are requested by the driver. It does not necessarily correspond to the load or the driving intention, and is not appropriate as a control parameter for setting the required fuel injection amount.

On the other hand, since the throttle valve opening changes in conjunction with the amount of depression of the throttle petal by the driver, the load and the driving intention requested by the driver are reflected, and the throttle valve opening and engine speed are It is conceivable to set the basic value of the fuel injection amount from the fuel amount setting map in accordance with the above. However, although the throttle valve opening reflects the driver's driving intention, it only indirectly corresponds to the air flow rate. Therefore, for example, each time the design of the throttle body is changed, the fuel amount setting map is changed. Need to be reset.

Moreover, the above-described fuel amount setting map corresponds to the late injection mode and the early injection mode, and in the late injection mode, there are operating modes in which EGR is performed and modes in which the EGR is not performed. The fuel amount setting map must be switched according to the above. In addition, if it is the first term injection mode, there is also lean combustion mode, theoretical air-fuel ratio feedback mode, open loop mode,
It is necessary to prepare a large number of fuel amount setting maps according to the operating mode of the engine, such as the operating mode in which EGR (exhaust gas recirculation) is performed and the mode in which it is not. In addition to the fuel injection amount, the basic value of ignition timing and EG
Many maps are required for each engine operating mode to set the R amount. Such a large number of fuel amount setting maps, ignition timing setting maps, EGR amount setting maps, etc.
If it is experimentally reset for each design of the throttle body and the like, its labor and cost will be enormous.

The cylinder injection type engine has more modes to be switched than the intake pipe injection type engine, and as described above, the pressure in the intake passage and the intake air amount which are directly related to the load cannot be used as parameters. So it will be a particularly important issue. The present invention has been made in view of the above problems, and a setting map for setting a parameter value that affects the combustion state in a cylinder, such as a fuel injection amount and an ignition timing, is provided with time and labor for each engine design change. An object of the present invention is to provide a control device for a cylinder injection type spark ignition type internal combustion engine which does not require spending and resetting.

[0011]

Therefore, according to claim 1 of the present invention, at least the engine speed and load information are detected in a control device for a direct injection type spark ignition internal combustion engine that injects fuel directly into a combustion chamber. Operating state detecting means, calculating means for calculating target average effective pressure information based on the detection result of the operating state detecting means, and combustion state in the combustion chamber according to the target average effective pressure information calculated by the calculating means. And a combustion parameter setting means for setting a parameter value that affects the internal combustion engine based on the parameter value set by the combustion parameter setting means, a cylinder injection type spark ignition internal combustion engine A controller is provided.

Here, the average effective pressure (MPa) is one of the indexes showing the engine output performance, and generally, assuming that the absorption torque T obtained from the dynamometer, the average effective pressure Pe = 4πT / Ve (4 Cycle engine). Here, Ve is the stroke volume. Also,
The effective work We obtained by subtracting the friction loss from the positive work of the engine obtained from the acupressure diagram may be divided by the stroke volume Ve of the engine (= We / Ve), which is called the indicated average effective pressure. . The target average effective pressure information includes the indicated average effective pressure, the net average effective pressure after subtracting mechanical loss and heat loss, the net output, and the like.

The parameter value that affects the combustion state may be a fuel injection amount, a fuel injection end timing, an ignition timing, an exhaust gas recirculation amount that recirculates exhaust gas to the intake side, or the like. According to the third aspect of the invention, the switching means is provided for switching between the first injection mode in which fuel injection is mainly performed in the intake stroke and the latter injection mode in which fuel injection is mainly performed in the compression stroke according to the detection result of the operating state detection means. In the invention of claim 4, the calculating means is used when the mode is switched to the first injection mode, and is used when the first calculation map for calculating the target average effective pressure information and the latter injection mode is used. Second for calculating target average effective pressure information
And a calculation map.

The target average effective pressure information calculated from the second calculation map may be calculated at least according to the engine speed and the throttle valve opening (claim 5). On the other hand, the first
The target average effective pressure information calculated from the calculation map may be calculated at least according to the engine speed and the pressure in the intake passage downstream of the throttle valve (Claim 7), or at least the engine speed and the fresh air intake air. It may be calculated according to the amount (claim 8). Further, the target average effective pressure information calculated from the first calculation map may be calculated at least according to the engine speed and the volumetric efficiency (Claim 9). In this case, the volumetric efficiency is the fresh air intake air amount. And a pressure in the intake passage downstream of the throttle valve (claim 10).

According to a sixth aspect of the present invention, when the first injection mode is switched to the second injection mode by the switching means,
The calculation means is configured to calculate the target average effective pressure information by using the second calculation map at the same time as switching.
According to the first aspect of the present invention, when the switching means switches the late injection mode to the early injection mode, the calculating means calculates the target average effective pressure information using the first calculation map after a predetermined period has elapsed from the time of switching. Is composed of.

[0016]

In the control device for the internal combustion engine according to the first aspect, the target average effective pressure information is calculated based on at least the engine speed and the load information, and the fuel injection amount is calculated according to the calculated target average effective pressure information. , Ignition timing, exhaust gas recirculation amount (claim 2), and other parameter values that affect the combustion state in the combustion chamber are set. Therefore, if the average effective pressure information and the combustion parameter value are associated with each other in advance, only the relationship between the throttle valve opening and the average effective pressure information is associated even if the passage area of the throttle body is changed. It will be good.

According to the third aspect of the present invention, there is provided a switching means for switching the control of the engine between the early injection mode and the late injection mode. Even in such an engine, the target average effective pressure information depends on the switched mode. By performing the calculation, the combustion parameter setting means need only set the combustion parameter value according to the calculated target average effective pressure information.

The target average effective pressure information set by the calculating means is suitable for computer control by using the first and second calculation maps prepared in advance for each mode (claim 4). The late injection mode is switched during low load operation such as idle operation, and from the second calculation map used in the late injection mode, the throttle valve opening and engine speed that represent the driver's driving intention are set. Accordingly, the target average effective pressure information is calculated (Claim 5). On the other hand, normally, during the operation in which the throttle valve is opened, the mode is switched to the first term injection mode, and from the first calculation map used in the first term injection mode, the intake passage pressure (claim 7) and the fresh air intake air amount are calculated. (Claim 8)
And target average effective pressure information is calculated according to the engine speed.

When the throttle valve is opened and the late injection mode is switched to the early injection mode, a response delay occurs until the intake passage pressure or the fresh air intake air amount reaches a value corresponding to the throttle valve opening. The calculation map for calculating the target average effective pressure information is switched after waiting for the period corresponding to the response delay period. on the other hand,
When the early injection mode is switched to the late injection mode, the frothle valve opening itself represents the driving intention of the driver, so that the map is switched to a calculation map for calculating the target average effective pressure information without a time delay (claim 6). .

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a schematic configuration diagram showing an embodiment of an engine control system to which the present invention is applied, and FIG. 2 is a vertical sectional view of a cylinder injection gasoline engine according to the embodiment. In these figures, reference numeral 1 denotes an in-cylinder in-cylinder in-line four-cylinder gasoline engine for an automobile (hereinafter simply referred to as an engine), in which a combustion chamber, an intake device, an EGR device and the like are designed exclusively for in-cylinder injection. .

In the case of the present embodiment, the cylinder head 2 of the engine 1 is provided with an electromagnetic fuel injection valve 4 together with a spark plug 3 for each cylinder so that the fuel is directly injected into the combustion chamber 5. It is like this. A hemispherical cavity 8 is formed on the top surface of the piston 7 that reciprocates by sliding in the cylinder 6, at a position where fuel spray from the fuel injection valve 4 reaches near the top dead center. (Figure 2). Also,
The theoretical compression ratio of the engine 1 is set higher (about 12 in this embodiment) than that of the intake pipe injection type. A DOHC four-valve system is adopted as a valve mechanism, and an intake side camshaft 11 and an exhaust side camshaft 12 are rotatably provided in an upper portion of the cylinder head 2 so as to drive the intake and exhaust valves 9 and 10, respectively. Is held.

Both camshafts 1 are attached to the cylinder head 2.
An intake port 13 is formed in a substantially upright direction so as to pass through between 1 and 12, and the intake flow passing through the intake port 13 generates a reverse tumble flow in the combustion chamber 5 described later. ing. On the other hand, the exhaust port 14 is formed in a substantially horizontal direction like a normal engine, but an EGR port 15 (not shown in FIG. 2) having a large diameter is obliquely branched. In the figure, 16 is a water temperature sensor for detecting the cooling water temperature TW, and 17 is a crank angle sensor for outputting a crank angle signal SGT at a predetermined crank position of each cylinder (5 ° BTDC and 75 ° BTDC in this embodiment). And 19 is an ignition coil that outputs a high voltage to the ignition plug 3. A cylinder discriminating sensor (not shown) that outputs a cylinder discriminating signal SGC is attached to a camshaft or the like that rotates at a half speed of the crankshaft to discriminate which cylinder the crank angle signal SGT belongs to. .

As shown in FIG. 2, the intake port 13 includes an air cleaner 22, a throttle body 23, and a stepper motor type ISCV (idle speed control valve) 24 via an intake manifold 21 having a surge tank 20. The intake pipe 25 is connected.
Further, the intake pipe 25 is provided with a large-diameter air bypass pipe 26 that bypasses the throttle body 23 and introduces intake air into the intake manifold 21, and a linear solenoid type large-sized ABV is provided in the pipe line. An (air bypass valve) 27 is provided. In addition, the air bypass pipe 2
6 has a flow passage area corresponding to that of the intake pipe 25.
When the V27 is fully opened, the required amount of intake air can be circulated in the low and medium speed range of the engine 1. On the other hand, ISCV2
No. 4 has a flow passage area smaller than that of the ABV 27, and the ISCV 24 is used when the intake air amount is accurately adjusted.

The throttle body 23 is provided with a butterfly type throttle valve 28 for opening and closing a flow path, a throttle sensor 29 for detecting an opening θTH of the throttle valve 28, and an idle switch 30 for detecting a fully closed state. ing. In the figure, 31 is a boost pressure (MAP: Manifold Absolute Pressure) for detecting the intake pipe pressure Pb.
It is a sensor and is connected to the surge tank 20.

On the other hand, the exhaust port 14 has an O ₂ sensor 4
An exhaust pipe 43 including a three-way catalyst 42 and a muffler (not shown) is connected via an exhaust manifold 41 to which 0 is attached. In addition, the EGR port 15 has a large diameter EG.
Downstream of the throttle valve 28 via the R pipe 44,
Moreover, it is connected to the upstream side of the intake manifold 21,
A stepper motor type EGR valve 45 is provided in the conduit.

The fuel tank 50 is installed in the rear portion of the vehicle body (not shown). Then, the fuel stored in the fuel tank 50 is sucked up by the electric low-pressure fuel pump 51,
It is fed to the engine 1 side via the low-pressure feed pipe 52. The fuel pressure in the low-pressure feed pipe 52 is the first fuel pressure regulator 5 installed in the conduit of the return pipe 53.
4, the pressure is adjusted to a relatively low pressure (3.0 kg / mm ^{2 in} this embodiment, hereinafter referred to as low fuel pressure). The fuel fed to the engine 1 side is fed to each fuel injection valve 4 by a high-pressure fuel pump 55 attached to the cylinder head 2 via a high-pressure feed pipe 56 and a delivery pipe 57. In the case of the present embodiment, the high-pressure fuel pump 55 is of the swash plate axial piston type, is driven by the exhaust side cam shaft 12, and is 50 kg / mm ² even when the engine 1 is idle.
The above discharge pressure is generated. The fuel pressure in the delivery pipe 57 is set to a relatively high pressure (50 in the present embodiment) by the second fuel pressure regulator 59 provided in the conduit of the return pipe 58.
kg / mm ² . Hereinafter, it will be described as high fuel pressure). In the figure,
Reference numeral 60 denotes an electromagnetic fuel pressure switching valve attached to the second fuel pressure regulator 59, which relieves fuel in the ON state and
The fuel pressure in the delivery pipe 57 is set to a predetermined value (for example, 3.0 kg /
mm ² ). Further, 61 is a high-pressure fuel pump 55.
Is a return pipe for returning the fuel, which has been lubricated and cooled, to the fuel tank 50.

An ECU (electronic control unit) is provided in the passenger compartment.
70 is installed in the ECU 70, and a storage device (ROM, RAM, non-volatile RAM) for storing an input / output device (not shown), a control program, a control map, etc.
Etc.), a central processing unit (CPU), a timer counter, etc., and controls the engine 1 as a whole. EC
On the input side of U70, switches for detecting the operating conditions of the air conditioner device, the power steering device, the automatic transmission, etc., which become the load of the engine 1 during operation, that is, the air conditioner switch (A / C / SW) 33, the power steering device. A switch (P / S / SW) 34, an inhibitor switch (INH / SW) 35, etc. are connected (see FIG. 6),
Each detection signal is supplied to the ECU 70. The ECU 7
In 0, in addition to the various sensors and switches described above,
A large number of switches and sensors not shown are connected to the input side, and various warning lights and devices are also connected to the output side.

The ECU 70 determines the ignition timing, the amount of EGR gas introduced, etc., including the fuel injection mode and the fuel injection amount, based on the input signals from the various sensors and switches described above, and the fuel injection valve 4, Ignition coil 19, EGR
The valve 45 and the like are drive-controlled. Next, a basic flow of engine control will be briefly described.

When the driver turns on the ignition key in the cold state, the ECU 70 turns on the low-pressure fuel pump 51 and the fuel pressure switching valve 60 to supply the fuel injection valve 4 with fuel of low fuel pressure. This is because the high-pressure fuel pump 55 operates at all or only incompletely when the engine 1 is stopped or cranking.
This is because the fuel injection amount has to be determined based on the discharge pressure and the valve opening time of the fuel injection valve 4. Next, when the driver starts the ignition key, the engine 1 is cranked by a starter motor (not shown), and at the same time fuel injection control by the ECU 70 is started. At this time, the ECU 70 selects the first-stage injection mode and injects fuel so that the air-fuel ratio becomes relatively rich. This is because the vaporization rate of the fuel is low when the engine is cold, so that misfire and discharge of unburned fuel (HC) cannot be avoided when injection is performed in the late injection mode (that is, the compression stroke). Further, since the ECU 70 closes the ABV 27 at the time of starting, the intake air to the combustion chamber 5 is supplied from the gap of the throttle valve 28 and the ISCV 24. The ISCV 24 and the ABV 27 are centrally managed by the ECU 70,
Each valve opening amount is determined according to the required introduction amount of intake air (bypass air) that bypasses the throttle valve 28.

When the engine 1 starts the idle operation after the start-up is completed, the high pressure fuel pump 55 starts the discharge operation of the rated value. To supply. At this time, of course, the fuel injection amount is determined based on the high fuel pressure and the valve opening time of the fuel injection valve 4. Then, until the cooling water temperature TW rises to a predetermined value, the ECU 70 selects the first-term injection mode to inject fuel as at the time of starting, and continuously closes the ABV 27 as well. Further, the control of the idle speed according to the increase or decrease of the load of the auxiliary equipment such as the air conditioner is performed by the ISCV 24 (the ABV 27 is also opened if necessary) as in the intake pipe injection type. Furthermore,
When the O ₂ sensor 40 reaches the activation temperature after the lapse of a predetermined cycle, the ECU 70 starts the air-fuel ratio feedback control according to the output voltage of the O ₂ sensor 40 and purifies the harmful exhaust gas component by the three-way catalyst 42. . As described above, in the cold state, the fuel injection control which is substantially the same as that of the intake pipe injection type is performed, but since the fuel droplets are not attached to the wall surface of the intake pipe 13, the control response and accuracy are improved. .

When the engine 1 has finished warming up, the ECU 7
0 is based on the in-cylinder effective pressure (target average effective pressure) Pe obtained from the intake pipe pressure Pb, the throttle opening θTH, etc. and the engine rotation speed Ne, based on the fuel injection control map of FIG. To determine the fuel injection mode and the fuel injection amount to drive the fuel injection valve 4, and
Also, the opening control of the EGR valve 45 and the like are performed.

For example, during low load / low speed operation such as during idle operation, the late injection lean region in FIG.
The CU 70 selects the late injection mode (also referred to as the late lean mode) and opens the ABV 27 and the EGR valve 40 in accordance with the operating state to obtain a lean air-fuel ratio (about 20 to 40 in this embodiment). So that the fuel is injected. At this point, the fuel vaporization rate increases and
As shown in FIG. 5, the intake flow flowing from the intake port 13 forms a reverse tumble flow 80 indicated by an arrow, so that the fuel spray 8
1 is stored in the cavity 8 of the piston 7. As a result, an air-fuel mixture near the stoichiometric air-fuel ratio is formed around the spark plug 3 at the time of ignition, and it is possible to ignite even with an extremely lean air-fuel ratio (for example, the total air-fuel ratio is about 40). Become. As a result, the amount of CO and HC emitted is extremely small, and NOx is generated by the exhaust gas recirculation.
Emissions of can be kept low. And ABV27 and E
The fuel consumption is greatly improved in combination with the reduction of pumping loss by opening the GR valve 40. The control of the idle speed according to the increase / decrease of the load is performed by increasing / decreasing the fuel injection amount, so that the control responsiveness becomes very high.

In the latter injection mode, the injection valve 4
The fuel spray injected from the vehicle must reach the spark plug 3 by riding on the above-described reverse tumble flow, and the fuel must be vaporized by the time it reaches the ignition point to form an air-fuel mixture that facilitates ignition. I won't. When the average air-fuel ratio becomes 20 or less, an overrich air-fuel mixture is locally generated near the spark plug 3 to cause so-called rich misfire, while when it becomes 40 or more, it exceeds the lean limit and misfire (so-called lean misfire) also occurs. . Therefore, the timing of fuel injection start and end is accurately controlled, and the average air-fuel ratio is 20 to 40.
Is set to fall within this range, and if it exceeds this range, it is switched to the later-described injection mode or the like.

Further, during low-medium speed running, depending on the load condition and the engine speed Ne, the previous period injection lean region or stoichio feedback region (theoretical air-fuel ratio feedback control region; / B range), the ECU 70 determines whether the lean mode or S-F
The / B mode (these two modes and the open loop control mode to be described later are collectively referred to as the first term injection mode) is selected, and the fuel is injected so as to have a predetermined air-fuel ratio.

That is, in the early lean mode, the valve opening amount and the fuel injection amount of the ABV 27 are controlled so that the air-fuel ratio becomes relatively lean (about 20 to 23 in this embodiment), and S-
In the F / B mode, the open / close control of the ABV 27 and the EGR valve 45 is performed, and the air-fuel ratio feedback control is performed according to the output voltage of the O ₂ sensor 40. Also in this case, FIG.
As shown in FIG. 5, the intake flow that flows in from the intake port 13 forms the reverse tumble flow 80. Therefore, by adjusting the fuel injection disclosure timing or the end timing, the lean turbulence causes lean turbulence even in the early injection lean region. It is possible to ignite at various air-fuel ratios. Note that the ECU 70 also opens the EGR valve 45 in this control region so that an appropriate amount of E can be stored in the combustion chamber 5.
By introducing the GR gas, NO _X generated at a lean air-fuel ratio is significantly reduced. Also, S-F / B
In the region, a large output is obtained due to the relatively high compression ratio, and harmful exhaust gas components are purified by the three-way catalyst 42.

Since the open-loop control range in FIG. 3 is set at the time of sudden acceleration or high-speed running, the ECU 70 selects the previous injection mode and closes the ABV 27, throttle opening θTH, engine speed Ne, etc. According to the above, the fuel is injected so that the air-fuel ratio becomes relatively rich. At this time, the compression ratio is high and the intake flow is the reverse tumble flow 80.
In addition to the above, the intake port 13 is substantially upright with respect to the combustion chamber 5, so a high output can be obtained due to the inertia effect.

Further, during coasting operation during medium-high speed traveling, the fuel cut region in FIG. 3 is reached, so the ECU 70 completely stops fuel injection. As a result, the fuel consumption is improved and at the same time, the emission amount of the harmful exhaust gas component is reduced. The fuel cut is immediately stopped when the engine rotation speed Ne becomes lower than the return rotation speed or when the driver depresses the accelerator pedal.

Next, according to the present invention, a parameter value that is set by the target average effective pressure information and affects the combustion state in the engine combustion chamber, that is, the valve opening time Tinj of the fuel injection valve 4, the ignition timing Tig, and the EGR. The procedure for setting the valve opening amount Legr, etc. of the valve 45 will be explained, and between the late lean mode and the SF / B mode, the early lean mode and the SF.
Taking the mode switching between the / B mode and between the first lean mode and the second lean mode as an example, the control procedure at the time of switching the modes will be described.

In FIG. 6, the target average effective pressure Pe is calculated,
FIG. 26 is a block diagram showing a procedure for calculating a target A / F, an injection end timing Tend, a basic ignition timing θB, a valve opening degree Leg of the EGR valve 45 and the like according to the target average effective pressure Pe, and FIGS. 6 is a flowchart showing a control procedure for determining an engine control mode and shifting to the mode, and a control procedure in that mode. Therefore, the engine control procedure of the present invention will be sequentially described by following this flowchart. The combustion parameter setting routine shown in FIGS. 14 to 22 is executed every time the ECU 70 detects a predetermined crank angle position of each cylinder.

First, the ECU 70 determines and sets the control mode in steps S1 to S8 shown in FIG. The details of the control in the control mode to be executed have been described with reference to FIG. 3, so a detailed description thereof will be omitted, but the control mode to be executed is determined based on the detection information from various sensors and switches. To be done. Then, for example, when the late lean mode is determined in step S1 (the determination result of step S1 is affirmative (Yes
In the case of s)), various control flags and control variables are set in step S2 to execute the control in the late lean mode. When the previous lean mode is determined in step S5 (when the determination result in step S5 is affirmative), various control flags and control variables are set in step S6 to execute the control in the previous lean mode.

However, as described above, step S2
Also, even if the control flags for the late lean mode and the early lean mode are set in step S6, the ECU 70 executes step S4 and determines whether the engine 1 is accelerating. Whether or not the engine 1 is accelerating is determined by the deviation (time change rate) Δθ between the previous value and the current value of the throttle valve opening θth detected by the throttle sensor 29, and the engine speed detected by the crank angle sensor 17. The difference between the previous value and the current value of Ne (the time change rate of the engine speed) ΔN is discriminated. If the deviation Δθ or ΔN exceeds the respective prescribed discrimination values (α, β), it is determined to be acceleration, and acceleration is performed once. After it is determined that the deviation Δθ or ΔN is the predetermined discriminant value (α-Δα,
When it is less than β-Δβ), it is determined that the acceleration is completed. here,
(Δα, Δβ) are minute values for giving a hysteresis characteristic to stabilize the control, and these values can be set to appropriate values including 0.

If the determination result is affirmative in step S4 and it is determined that the engine 1 is accelerating, the process proceeds to step S8, and setting is made to forcibly execute the acceleration control in the S-F / B mode. The various control flags and control variables that have been used are changed to those according to the SF / B mode. Then, as long as the above acceleration condition is satisfied, this step S8 is repeatedly executed to perform the acceleration control. The acceleration control method in the S-F / B mode is not particularly limited, and a conventional acceleration control method can be used.
Note that once acceleration is determined, S-F /
It is also possible to execute the control in the B mode and continue the S-F / B mode control even if the acceleration release condition is satisfied during a predetermined period. This stabilizes control and improves drivability.

When the determination result of step S4 is negative (No), that is, when the acceleration state of the engine 1 is not detected or it is determined that the acceleration is completed, the control flag set in step S2 or step S6 is changed. The control is performed in the determined mode without any change. Neither the late lean mode nor the early lean mode (when both Step S1 and Step S5 are No, it is determined to be the early S-F / B mode, and the process proceeds to the above-mentioned Step S8 to perform the S-F / B mode. Set various control flags and control variables of.

Incidentally, in steps S2, S6 and S8, tailing coefficients K1, K2, KS and KL which will be described later are provided.
As shown in FIG. 26, the tailing coefficient corresponding to the mode transition mode is set to a value of 0 when the mode transition is discriminated. Each coefficient value is unchanged. For example, S-
When the transition from the F / B mode to the late lean mode is first determined, the tailing coefficient K1 is reset to 0. Further, the tailing coefficient KL is reset to a value of 0 when the transition from the lean mode to the SF / B mode is first discriminated.

When the setting of the various control flags and the like is completed, the ECU 70 proceeds to step S10 and subsequent steps in FIG. 15 to execute the transition control of each mode and the control in that mode. 27 to 29
This will be described with reference to the timing chart showing the change over time of various control parameter values. First, for convenience of explanation, starting from the case where the late lean mode is executed (assuming that the tailing coefficient value K1 is set to the value 1.0) (before time t0 in FIG. 27), the ECU 70 determines in step S10. Then, it is determined whether the mode to be controlled is the late mode or the early mode. The late mode means the late lean mode, and the early mode includes the early lean mode and the SF mode. Since the current operation mode of the engine 1 is the late lean mode as described above, step S12
, Various parameter values Pe necessary for engine control,
Kaf, Tig, Tend, Legr, Ev, etc. are calculated. The method of calculating these parameter values will be described below with reference to the block diagram of FIG. If the ECU 70 determines that the engine 1 is in the operating state in which the late lean mode control should be executed, the changeover switches (switching means) 70a and 70b shown in FIG. 6 are switched to the late mode side.

First, starting with the explanation of the calculation of various variable values related to the valve opening time Tinj of the fuel injection valve 4, the ECU 7
0 is the throttle valve opening θth and the engine rotation detected by the throttle sensor 29 and the crank angle sensor 17 from the target average effective pressure map (first calculation map of the calculation means) 70c stored in advance in the storage device. A target average effective pressure Pe is calculated according to the number Ne. Figure 7
Shows the details of the target average effective pressure map, and the target average effective pressure Peij corresponding to the output required by the driver in accordance with the throttle valve opening θth and the engine speed Ne is mapped to the storage device of the ECU 70. Remembered Each of these data is a value that is experimentally set by using the net average effective pressure, which is easy to collect in the bench test of the engine as the target average effective pressure information. From this map, the ECU 70 calculates the optimum target average effective pressure Pe according to the detected throttle valve opening θth and the engine speed Ne, for example, by a known 4-point interpolation method.

In this embodiment, the net average effective pressure Pe is used as the target average effective pressure information, but various types can be used if there is no particular problem in collecting data in the engine bench test. Alternatively, the indicated mean effective pressure or net output may be used. The storage device of the ECU 70 includes various load devices, such as an air conditioner device and a power steering device, which serve as mechanical and electrical loads of the engine 1 during operation.
Output correction maps 70d to 70f for the transmission and the like are provided, and target average effective pressure correction values corresponding to the engine speed Ne are output by ON signals from switches 33 to 35 that detect the operation of these load devices. To be done. These correction values are added to the target average effective pressure Pe obtained from the map 70c by the adder 70g to correct the value.

The target average effective pressure Pe calculated in this way is filtered by the first-order lag filter 70h and sent to the target air-fuel ratio correction coefficient value Kaf calculation map 70j which is the combustion parameter setting means. The reason for providing the first-order lag element (filter) 70h is that, when in-cylinder fuel injection is performed, a change in the injection amount immediately appears as a change in output or the like. On the other hand, the throttle valve opening θth that determines the fuel injection amount is detection information that can be detected without delay compared to the detection of the intake air amount and the like, and the detected valve opening θth
When the fuel injection amount corresponding to is immediately supplied to the engine 1,
Drivability may be impaired. Note that the first-order delay element 70h may not be provided in some cases, such as when prioritizing control responsiveness.

Details of the target air-fuel ratio correction coefficient value calculation map 70j are shown in FIG. 11, and a plurality of maps are prepared for each mode and corresponding to the presence or absence of EGR. Similar to the one shown in FIG. 7, it is experimentally set in advance according to the target average effective pressure Pe and the engine speed Ne, and is stored in the storage device described above. ECU
Reference numeral 70 denotes a target air-fuel ratio correction coefficient value calculation map 70j, which calculates a target air-fuel ratio correction coefficient value Kaf corresponding to the target average effective pressure Pe and the engine speed Ne input to the calculation map 70j, and opens the valve described later. Used for time calculation.

On the other hand, in the volumetric efficiency calculating means 70k, the volumetric efficiency Ev is determined in accordance with the target average effective pressure Pe and the engine speed Ne which are subjected to the first-order lag filtering as described above.
The value is calculated. FIG. 9 shows a volumetric efficiency map used in the latter lean mode control, and the volumetric efficiency map value shown in this map also depends on the target average effective pressure Pe and the engine speed Ne, as shown in FIG. Are experimentally set in advance and stored in the above-mentioned storage device.

The target air-fuel ratio correction coefficient value Kaf and the volumetric efficiency Ev obtained as described above are applied to the following equation (F1), and the valve opening time Tin of the fuel injection valve 4 is applied at the timing described later.
j is calculated. Tinj = K * Pb * Ev * Kaf * (Kwt * Kat * ...) * Kg + TDEC ... (F1) where Pb is the intake pipe pressure (pressure in the intake passage) detected by the boost pressure sensor 31. ), And Kwt, Ka
etc. are various correction factors set according to the engine water temperature Tw, the atmospheric temperature Tat, the atmospheric pressure Tap, and the like. Kg
Is a gain correction coefficient of the injection valve 4, TDEC is a dead time correction value, and the target average effective pressure Pe and the engine speed Ne are
And is set accordingly. K is a conversion coefficient for converting the fuel amount into the valve opening time, which is a constant.

Kaf is set according to the engine operating state. Note that the formula (F1) is applied to other modes regardless of the latter lean mode control, and among the various correction coefficients described above, the air-fuel ratio correction coefficient value Kaf is the latter lean mode and SF / It is set by a method described later when switching between modes such as B mode, set according to the output voltage of the O ₂ sensor 40 during S-F / B mode control, and set to a value optimal for that mode in other modes as well. To be done.
Further, it goes without saying that the value set in each mode is used as the volume efficiency Ev.

The volumetric efficiency Ev used to calculate the valve opening time Tinj in the above formula (F1) is supplied to each combustion chamber 5 and is related to combustion per unit intake stroke (per cylinder). It is an index related to the amount of oxygen that can be generated, and similar indices include filling efficiency, intake efficiency, etc., and volumetric efficiency E
These indexes can be used instead of v. The value obtained by the volumetric efficiency Ev and the intake pipe pressure Pb is
It is related to the intake air amount per unit intake stroke, and instead of using the volumetric efficiency Ev or the intake pipe pressure Pb, the intake air per unit intake stroke is obtained directly by the air flow rate detected by the air flow sensor and the engine speed. Amount (A / N) can be used. These volumetric efficiency, filling efficiency, intake air amount (A / N) per unit intake stroke, etc. are collectively referred to as an effective intake parameter.

The valve opening time Tinj thus calculated is sent to an injector drive circuit (not shown) for driving the fuel injection valve 4 at a predetermined timing. Next, the setting of the injection end timing Tend will be described.
In the injection end timing setting means (combustion parameter setting means) 70m shown in, the injection end timing Tend suitable for the control mode is set according to the target average effective pressure Pe and the engine speed Ne. If the injection end timing of the fuel injection in the late lean mode is delayed, a period for sufficiently injecting the injected fuel spray cannot be secured, and black smoke is generated. On the other hand, if it is too early, the injected fuel may collide with the cylinder wall or the like, so that an optimum mixture may not be formed, resulting in misfire. The injection end timing Tend is experimentally set in advance to an optimum value and mapped for each control mode or according to the presence or absence of EGR or the like. The injection end timing Tend set according to the target average effective pressure Pe and the like is further corrected by the engine water temperature and the like and supplied to the injector drive circuit. In the injector drive circuit, the injection start timing is calculated based on the supplied injection end timing Tend and the valve opening time Tinj, and when the calculated injection start timing comes, the valve opening time T for the fuel injection valve 4 of the cylinder to be injected is calculated.
The drive signal is output for the period corresponding to inj.

The ignition timing Tig is calculated by the ECU 70 as follows:
It is calculated based on (F2). Tig = θB + (various retard correction amounts) (F2) The basic ignition timing θB in the above equation is calculated by the ignition timing setting means (combustion parameter setting means) 70n in FIG. FIG. 12 schematically shows the configuration of the ignition timing setting means 70n, and has a plurality of basic ignition timing setting maps for each mode and for each operating state such as the presence or absence of EGR. During the latter lean mode control, the target average effective pressure Pe set in accordance with the above-mentioned target average effective pressure map 70c according to the throttle valve opening degree θth is supplied to the ignition timing setting means 70n, and the target average effective pressure Pe and Engine speed N
The basic ignition timing θB corresponding to e is calculated from the late lean mode map.

In addition to the ordinary correction values such as the engine water temperature correction value, the various retard amounts include the late lean mode and SF / F /
At the time of transition between the B modes, retard correction values at transition R1 (K) and R2 (K), which will be described later, are included. The retard correction values R1 (K) and R2 (K) at the time of transition are set to the value 0 except at the time of transition. The ignition timing in the latter lean mode control is set at the time when the optimum mixture reaches the spark plug 3, and this set timing becomes the optimum ignition timing.

The ignition timing Tig set as described above is supplied to an ignition coil drive circuit (not shown), and at a time corresponding to the set ignition timing Tig, a high voltage is applied to the ignition plug 3 of the cylinder to be ignited. Is applied to ignite. The valve opening degree Leg of the EGR valve 45 is calculated by the EGR amount setting means (combustion parameter setting means) 70p in FIG. FIG.
Reference numeral 3 shows a schematic configuration of the EGR amount setting means 70p. A plurality of EGR valves are provided for each operation mode in which exhaust gas is to be recirculated and for the selected position of the transmission (D range or N range). It has an opening map. In calculating the valve opening degree Legr, the target average effective pressure map 70c described above is used.
Therefore, the target average effective pressure Pe set according to the throttle valve opening θth is not subjected to the first-order lag filtering process, and the set target average effective pressure Pe is simply changed to the low-pass filter 70q.
Is supplied to the EGR amount setting means 70p via a valve opening degree L according to the target average effective pressure Pe and the engine speed Ne.
egr is calculated from the late lean mode map.

When the exhaust gas is supplied to the engine 1 through the EGR valve 45, E corresponding to the changed valve opening degree is obtained.
A large time lag occurs in supplying the GR amount to the engine 1. In consideration of such a time lag, it is better to calculate the EGR amount most suitable for the operating state, so that the target average effective pressure Pe set in the target average effective pressure map 70c is set.
Is supplied to the EGR amount setting means 70p without delay.

The valve opening degree Legr calculated as described above
Is configured to output to the EGR valve 45 a valve drive signal corresponding to the valve opening degree Legr after being supplied to an EGR drive circuit (not shown) after correction such as engine water temperature correction. Returning to step S12 in FIG. 15, when the calculation of various combustion parameter values and the like is completed as described above, the process proceeds to step S20 in FIG. In this step, it is judged whether or not the tailing coefficient K1 is 1.0. As described above, the tailing coefficient K1 is a value when the transition to the late lean mode is completed and the mode is controlled in that mode.
It is 1.0. Since the current engine operating state is controlled by the late lean mode after the complete transition, the tailing coefficient K1 is set to the value 1.0, and step S2 is performed.
Go to 1 and prepare for the transition from the latter mode to the first mode. To prepare for the transition, the initial values of the control variables for the transition are set, and various correction coefficient values Kaf and combustion parameter values Tig, Tend, EV, which are calculated in step S12 and are used in the current late lean mode control, The target average effective pressure Pe and the like are stored. Control variables for the transition include an invalid period counter Td2 and a boost pressure delay counter CNT2, and the former counter Td2 has a value f2 set as an initial value according to the target average effective pressure Pe and the engine speed Ne. (Ne, Pe), and the value XN2 is set in the latter counter CNT2. The stored values such as the initialization of the control variable value and the correction coefficient value Kaf are updated to new values every time this step S21 is executed.

When the setting of the initial values of the control variables and the like in step S21 is completed, the routine proceeds to step S22, where the late injection set routine is executed, and the various controls such as the fuel injection control, the ignition timing control and the EGR amount control described above are performed. To do. Next, the operating state of the engine 1 changes, and the engine is shifted from the late lean mode to SF.
Assuming that it is determined that the mode has shifted to / B mode,
In step S8 of FIG. 14 described above, the tailing coefficient K2 is set to the value 0 as shown in FIG. 26 (time t0 in FIG. 27). In such a case, the ECU 70
After determining the previous mode in step S10 of step S14, step S14 is executed, and various combustion parameter values Pe, Kaf, Tig, Tend, are executed in the same manner as step S12 described above.
Calculate Legr, Ev, etc.

In this case, the changeover switch 70 shown in FIG.
The a and 70b are switched to the previous term mode at the timing described later, and the target average effective pressure Pe is calculated by the target average effective pressure map 70r which is the second calculation map. In the previous lean mode or SF / B mode, the engine load required by the driver substantially corresponds to the intake pipe pressure Pb as in the normal intake pipe injection type, and the detected intake pipe pressure Pb itself is a primary factor. It has a delay element. Therefore, the intake pipe pressure Pb is used for setting the target average effective pressure Pe, since the first-order delay process is not required unlike the case where the target average effective pressure Pe is set with the throttle valve opening degree θth. Boost sensor 3
The intake pipe pressure Pb detected by 1 is supplied to the target average effective pressure map 70r, and the target average effective pressure Pe is calculated according to the intake pipe pressure Pb and the engine speed Ne. The method of calculating the target average effective pressure Pe is similar to that of the target average effective pressure map 70c.
A map similar to that shown in FIG. 7 as shown in FIG. 8 is prepared according to the engine operating condition such as the presence or absence of EGR.

In such a first-half mode, the intake air pressure Pb may be replaced by the fresh air intake air amount detected by the air flow sensor. Also, Engine 1
The volumetric efficiency Ev of the amount of air taken in by the engine is obtained based on the intake pipe pressure Pb or the fresh air intake air amount detected by the air flow sensor, and the target average effective pressure Pe is calculated according to the obtained volumetric efficiency Ev and the engine speed Ne. Can also be calculated.

When the target average effective pressure Pe is calculated, the target average effective pressure Pe is applied to the target air-fuel ratio correction coefficient value calculation map 70j, the injection end timing setting means 70m, the ignition timing setting means 70n, and the EGR amount setting means 70p. The target A / F and T are supplied using the maps according to the operating conditions.
end and Legr are calculated. Also, the volumetric efficiency calculation means 7
The intake pipe pressure Pb detected by the boost sensor 31 is also supplied to 0k, and the volumetric efficiency Ev is also shown in FIG. 10, using a map similar to that shown in FIG. 7, using the intake pipe pressure detection Pb and the engine speed Ne. The volumetric efficiency Ev is calculated according to
Also in this case, the intake pipe pressure Pb is used to calculate the volumetric efficiency Ev.
Instead of this, the fresh air intake air amount or the like detected by the air flow sensor may be used.

Then, the target A / obtained as described above
The valve opening time Tin of the fuel injection valve 4 is applied in the same manner as in the latter lean mode by applying F and the volumetric efficiency Ev to the previous equation (F1).
j is calculated. Returning to step S14 of FIG. 15, when the calculation of various combustion parameter values and the like is finished as described above,
It progresses to step S50 of FIG. In this step, it is determined whether or not the tailing coefficient K2 is 1.0. As described above, the tailing coefficient K2 is set to a value 0 immediately after the transition to the late lean mode,
Therefore, the determination result of step S50 is No, and the steps of step S51 and subsequent steps are executed to perform the transition process from the late lean mode to the SF mode. Although the tailing coefficient K2 becomes 1.0 when the transition processing is completed, until then, a small value ΔK2 smaller than 1.0 is sequentially added by the timer routine shown in FIGS. Until K2 reaches a value of 1.0, the transition processing according to the coefficient value K2 is performed.

23 and 24 show a flowchart of a timer routine executed by a clock pulse of a predetermined cycle generated by a built-in clock of the ECU 70. Various tailing coefficient values K1, K2, KL, KS are counted down by the clock pulse. The procedure is shown.
First, in steps S110 to S113, the tailing coefficient K1 is counted down. Coefficient value K1
Is added to a predetermined small value ΔK1 smaller than the value 1.0 (step S110), the coefficient value K1 is compared with the value 1.0 (step S112), and if it is larger than the value 1.0, it is reset to the value 1.0 (step S113). If the value is 1.0 or less, the process proceeds to the next step S114. Thus, once the tailing coefficient value is reset to 0, the small value ΔK1 is added every time this routine is executed, and if the added value reaches the value 1.0, it is held at that value. It has become.

The same applies to the other tailing coefficient values, and the tailing coefficient value K2 is determined in step S114.
Through countdown in step S117,
For the coefficient values KL and KS, similarly, step S118 is performed.
Through step S120 and step S122 through step S125, respectively. In addition,
The small values ΔK1, ΔK2, etc. added to each coefficient value determine the required length of the mode transition control period, and are set to different values for each tailing coefficient, but they may be set to the same value. it can.

In step S51, the invalid period counter T
It is determined whether or not d2 is 0, that is, whether or not the invalid period corresponding to the initial value f2 (Ne, Pe) of the counter Td2 has elapsed. Initial value f2 (Ne, P of counter Td2
e) is set by the execution of step S21 in FIG. 16 described above.
The counter value Td2 at the time when 1 is executed is equal to this initial value f2 (Ne, Pe). Therefore, step S51
The determination result of No becomes negative, the process proceeds to step S52, the predetermined value ΔTd2 is subtracted from the counter value Td2, and step S53 is performed.
In, the tailing coefficient value K2 is reset to 0. The steps S52 and S53 are repeatedly executed until the invalid period elapses, during which the tailing coefficient value K2 is held at the value 0. Here, the tailing coefficient K2 and the invalid period counter Td2 are both for avoiding a sudden change in the in-cylinder combustion state at the time of mode transition to improve drivability.

Next, the ECU 70 executes steps S55 and S57 to calculate the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency Ev by the following equations (F3) and (F4), respectively. Kaft = (1-K2) * Kaf '+ K2 * Kaf ... (F3) Ev = (1-K2) * Ev' + K2 * Ev ... (F4) where Kaf 'and Ev' are late lean. The target air-fuel ratio correction coefficient value and the volume efficiency calculated last during the mode control are stored as the Kaf 'value and the Ev' value when the above-described step S21 of FIG. 16 is executed last. Kaf and Ev of the last term on the right side of each equation are S
-Kaf is set when F / B mode control is executed.
The value is a value set according to the output value of the O ₂ sensor 40 (S
-Value calculated in the F / B mode).

Therefore, during the period in which the coefficient value K2 is 0 (invalid period from time t0 to time t1 shown in FIG. 27), the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency value Ev are the previous values,
That is, it is held at the last set value during the late lean mode control, but when the coefficient value K2 increases from 0 to 1.0, the coefficient value K2 becomes the value set by weighting according to the coefficient value K2. When it reaches 1.0, it is set to the value calculated in the SF mode. By the change of the tailing coefficient value K2 as described above,
As will be described later, the target air-fuel ratio correction coefficient value Kaf at the time of mode transition has a volume efficiency Ev of t1 as shown in FIG. 27 (g).
From the time point to the time point t4, the value is gradually changed linearly, and after the time point t4, the value is held at the value calculated by the SF / B mode.

Next, the ECU 70 executes step S in FIG.
In step 60, it is determined whether or not the boost pressure delay counter CNT2 has counted down to the value 0. If the boost pressure delay counter value CNT2 is not yet counted down to the value 0 (the determination result is negative), step S61.
27 to rewrite the target average effective pressure Pe to the value Pe ', and the last set value during the late lean control is the predetermined period (the period corresponding to the initial value XN2 of the counter, as shown in FIG. 27 (b)).
It is held for the period from time t0 to time t2 shown in (1). During the period corresponding to the initial value XN2, the throttle valve 28
Even if the valve is opened, the rise of boost pressure will be delayed, so
It is set in relation to this delay, and the initial value XN
2 is set for a predetermined number of strokes of the engine 1. The boost pressure delay counter value CNT2 delays the switching of the map for calculating the target average effective pressure Pe.

The count value of the counter CNT2 is
In the crank interrupt routine shown in FIG. 25, which is executed each time the predetermined crank angle position of each cylinder is detected, the value 1 is counted down. In this routine, in addition to the counter CNT2, the count values of the counters CNT1 and CNT3 are also counted down to 1. Step S6
When the determination result of 0 is affirmative (when the boost pressure delay period has elapsed), the following S-F / B mode control is performed as shown in FIG.
The value calculated from the target average effective pressure map 70r is used (after time t2 in FIG. 27).

Next, in step S62, the above equation (F3)
It is determined whether or not the temporary target air-fuel ratio correction coefficient value Kaft calculated in step 1 is smaller than the value Xaf. This discriminant value Xaf is a value at which there is a risk of rich misfire in the engine combustion chamber 5 when the engine is controlled in the late lean mode using the target air-fuel ratio correction coefficient value Kaf of this value. About 20
(Theoretical air-fuel ratio is 14.7). That is, if the target air-fuel ratio correction coefficient value Kaf is smaller than the value Xaf, it means that the engine output can be controlled by adjusting the fuel injection amount in the late lean mode. Until Kaft reaches the value Xaf (Fig. 27
Until time t3 shown in (c)), the target air-fuel ratio correction coefficient value Kaf is set to a value corresponding to the tailing coefficient K2, that is, a temporary target air-fuel ratio correction coefficient value Kaft (step S63). Then, in order to continuously execute the late lean mode control, the ignition timing Tig is held at the last value Tig 'set in the late lean mode (step S64), and the fuel injection end period Tend is also set to the last value set in the late lean mode. Value of Ten
It is held at d '(step S65).

Kaf = Kaft Tig = Tig 'Tend = Tend' After resetting the respective combustion parameter values in this way, step S22 in FIG. 16 described above is executed, and engine control in the late lean mode is performed.

When the tailing coefficient value K2 increases and the temporary target air-fuel ratio correction coefficient value Kaft exceeds the determination value Xaf, the determination result of step S62 becomes negative and the steps S63 to S65 described above are not executed. It proceeds to step S66. As a result, the target air-fuel ratio correction coefficient value K
The af and the fuel injection end period Tend are not rewritten to the earliest provisional target air-fuel ratio correction coefficient value Kaft and the final value Tend ′ calculated in the late lean mode, respectively, and the values calculated in the SF mode are unchanged. used. In this case, as can be seen from the change in the correction coefficient value at time t3 in FIG. 27 (c), the target air-fuel ratio correction coefficient value Kaf
Changes in a stepwise manner to a suitable value corresponding to a value in the vicinity of the stoichiometric air-fuel ratio in the S-F / B mode, and at this point, shifts to the S-F / B mode. That is, when the air-fuel ratio reaches a value (approximately 20) corresponding to the Xaf value of the rich misfire limit in the late lean mode control, the air-fuel ratio becomes the value 2
The intermediate value between 0 and the stoichiometric air-fuel ratio is not gradually changed, but is suddenly changed to near the stoichiometric air-fuel ratio in the SF / B mode. Along with this, the fuel injection end period Tend is also changed to a value suitable for the SF / B mode control (time t3 in FIG. 27 (d)).

In step S66, it is determined whether the previous mode is the previous lean mode or the SF / B mode, and different control is performed according to the determination result. Since the S-F / B mode is discriminated in the current loop (the discrimination result in step S66 is negative), step S67 is executed and the ignition timing Tig is replaced with a value calculated by the following equation (F5).

Tig = (1−K2) * Tig ′ + K2 * Tig + R2 (K2) ... (F5) where R2 (K2) is a retard set to prevent a sudden change in output due to mode transition. The quantity is shown in Figure 2.
As shown in the change from time t3 to time t4 of 7 (f), it takes a negative value temporarily as a function of the tailing coefficient value K2, and then gradually changes from that value at time t4 (K2 = 1.0). Set to the value 0. As a result, the ignition timing Tig is shown in FIG.
It changes as shown from time t3 to time t4 in (e).
By controlling the ignition timing Tig in this way, SF
A sudden increase in output due to the start of / B mode control is prevented.

After setting each combustion parameter value in this way, step S48 of FIG. 18 is executed, and engine control is performed in the first injection mode. Tailing coefficient value K
When 2 gradually increases and reaches the value 1.0, the determination result in step S50 of FIG. 19 becomes affirmative (Yes),
Step S58 is executed. This determination is to determine whether the previous mode control is the previous lean mode or the SF / B mode, and different control is executed according to the result. If the S-F / B mode is subsequently determined, the ECU 70 proceeds to step S70 of FIG. 21 and prepares for the late lean mode control transition or the early lean mode control transition. To prepare for the transition, the initial values of the control variables for the transition are set, and various correction coefficient values Kaf calculated in the current control mode, combustion parameter values Tig, Tend, EV, target average effective pressure Pe, etc. are stored. Keep it. As the control variables for the transition, there are an invalid period counter Td1 and an EGR delay counter CNT1, and the former counter Td1 has a value f2 (which is set as an initial value according to the target average effective pressure Pe and the engine speed Ne). Ne, P
e), the value XN1 is set in the latter counter CNT1. These transfer control variables, etc. are S-F / B
When the control according to the mode is repeated and step S70 is repeatedly executed, the new value is updated each time.

When the initial values of the control variables and the like have been set in step S70, the process proceeds to step S72, in which the tailing coefficient value KL used during the transition control from the lean mode to the SF / B mode mode is 1.0. It is determined whether or not there is. At present, the shift to the S-F / B mode has been completed and the control of that mode is being performed, so the coefficient value KL is 1.0
And skips step S73 and proceeds to step S74. In step S74, the count value of the EGR delay counter CNT3 described later is determined.
Even if NT3 is set to the initial value, since the countdown is always executed in the crank interrupt routine shown in FIG. 25 described above, the value is set except during the transition control from the lean mode to the SF mode, which will be described later. It should have been counted down to zero. After all, as long as the S-F / B mode after completion of the mode transition is executed, the determination result of step S74 is also negative, the process skips step S75 and proceeds to step S48 of FIG. To be executed. It should be noted that steps S73 and S75 will be described in transition control from the previous lean mode to the SF mode, which will be described later.

Next, the shift control when shifting from the current SF / B mode to the late lean mode again will be described. When the late lean mode is discriminated during the SF / B mode control in step S1 of FIG. 14 (FIG. 27 (a)).
At time t6), the tailing coefficient K1 in step S2
Is set to the value 0. Then, step S10 of FIG.
After that, the latter mode is determined, and after calculating various combustion parameter values and the like in step S12 described above, step S20 of FIG. 16 is executed to determine whether K1 is equal to 1.0. Immediately after the latter lean mode is determined, the tailing coefficient value K1 is 0 as described above, so the determination result in step S20 is negative, and the steps after step S24 are executed to change from the S-F / B mode to the latter lean mode. Perform the process of shifting to the mode. The tailing coefficient K
The value of 1 becomes 1.0 when the transition processing is completed, but until then, the minute routine ΔK1 smaller than 1.0 is sequentially added by the timer routine shown in FIGS. 23 and 24, and the tailing coefficient value K1 becomes 1.0. Until it becomes, the transition process according to the coefficient value K1 is performed.

In step S24, the invalid period counter T
It is determined whether or not d1 is the value 0, that is, whether or not the invalid period corresponding to the initial value f1 (Ne, Pe) of the counter Td1 has elapsed. Initial value f1 of counter Td1 (Ne, P
e) is set by the execution of step S70 in FIG. 21 at the time of the S / F / B mode control immediately before the shift, and the counter value Td1 at the time when this step S24 is performed immediately after the mode shift is Initial value f1 (Ne, P
equal to e). Therefore, the determination result in step S24 is negative, the process proceeds to step S25, the predetermined value ΔTd1 is subtracted from the counter value Td1, and the tailing coefficient value K1 is reset to 0 in step S26. Then, these steps S25 and S26 are repeatedly executed until the invalid period elapses (from time t6 to time t7 in FIG. 27 (c)), during which the tailing coefficient value K1 is held at 0. .

Next, the ECU 70 executes step S28 and step S30 of FIG. 17 to calculate the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency Ev by the following equations (F6) and (F7), respectively. Kaft = (1-K1) * Kaf '+ K1 * Kaf ... (F6) Ev = (1-K1) * Ev' + K1 * Ev ... (F7) The above formulas (F6) and (F7) are as described above. Which are similar to the equations (F3) and (F4), respectively, where Kaf 'and Ev' are the target air-fuel ratio correction coefficient value and the volume efficiency calculated last during the SF / B mode control, respectively. It is stored as the Kaf 'value and the Ev' value when the step S70 of 21 is finally executed. Kaf and Ev of the last term on the right side of each equation are values calculated in the latter lean mode this time.

Therefore, during the period in which the coefficient value K1 is 0 (invalid period from time t6 to time t7 shown in FIG. 27), the temporary target air-fuel ratio correction coefficient value Kaft and the volume efficiency value Ev are the previous values,
That is, it is held at the last set value during S-F / B mode control, but when the coefficient value K1 increases from 0 to 1.0, the value set by weighting according to the K1 value becomes
When the coefficient value K1 reaches the value 1.0, the coefficient values are set to the values calculated in the late lean mode.
Due to the change in the tailing coefficient value K1 as described above, the target air-fuel ratio correction coefficient value Kaf at the time of mode transition is, as will be described later, the volume efficiency Ev, as shown between the time points t7 and t10 in FIG. 27 (g). Gradually change its value linearly to t10
After the time point, the value is held at the value calculated by the late lean mode.

Next, the ECU 70 executes step S in FIG.
In step 31, it is determined whether or not the EGR delay counter CNT1 has counted down to the value 0. This counter CN
T1 is provided for the purpose of delaying the EGR control in the late lean mode, and prevents an excessive EGR state during the transition control from the SF mode to the late lean mode in which a large amount of EGR is introduced. EGR delay counter value CNT1
Is not yet reduced to the value 0, step S32 is executed to open the valve opening degree Leg of the EGR valve 45.
Rewrite r to the value Legr ', and set the value last set during S-F / B mode control for a predetermined period (the period corresponding to the initial value XN1 of the counter, from time t6 to time t9 shown in FIG. 27 (h). Period)). The period corresponding to the initial value XN1 is set in consideration of delaying the shift of the EGR amount to a value compatible with the late lean mode.

If the determination result of step S31 is affirmative (when the EGR delay period has elapsed), the aforementioned step S31 is performed.
32 is skipped, and the value calculated by the EGR amount setting means 70p of FIG. 6 is used for the subsequent lean mode control thereafter (after time t9 in FIG. 27). Next, in step S34, it is determined whether the temporary target air-fuel ratio correction coefficient value Kaft calculated by the above equation (F6) is smaller than the value Xaf. As described in step S62 of FIG. 20, the determination value Xaf is a value at which misfiring occurs when the fuel is rich in the late lean mode, or about 20 in terms of the air-fuel ratio (theoretical air-fuel ratio 14.
Although it is set to a value corresponding to 7), it is not necessary to set the same value as the value set in step S62 in some cases. If the target air-fuel ratio correction coefficient value Kaf is smaller than the value Xaf, it means that the engine output can be controlled by adjusting the fuel injection amount in the late lean mode, and the late lean mode is determined by the determination in step S34. Is to determine whether or not to start. If the target air-fuel ratio correction coefficient value Kaf is equal to or greater than the value Xaf, the SF / B mode control is continuously executed.

The ECU 70 determines whether the determination result of step S34 is negative, that is, the temporary target air-fuel ratio correction coefficient value K.
Until aft reaches the value Xaf (from time t7 to time t8 shown in FIG. 27 (c)), the process proceeds to step S40 in FIG. 18, where the injection end period Tend is the final value Tend calculated in the SF mode. Rewrite to 'and hold this value. And whether the control mode before the transition determination was the lean mode in the previous period,
In order to determine whether the mode is the S-F / B mode, it is determined whether or not the correction coefficient value Kaf 'set and stored immediately before the transition determination is smaller than the value 1.0. When the lean mode control is executed in the previous period, the correction coefficient Kaf is always set to a value smaller than 1.0.

If the determination result of step S42 is negative, that is, if the control mode before the transition determination is the SF / B mode, the transition determination of the target air-fuel ratio correction coefficient value Kaf is performed in step S46. The value just before being broken Kaf '
Is held. Then, the adjustment of the engine output by subsequently executing the S-F / B mode control is controlled by the ignition timing, the process proceeds to step S47, and the ignition timing Tig is set to a value calculated by the following formula (F5). replace.

Tig = (1−K1) * Tig ′ + K1 * Tig + R1 (K1) ... (F8) where R1 (K1) is a retard set to prevent a sudden change in output due to mode transition. The quantity is shown in Figure 2.
7 (f) is set to such a value that the retard amount gradually increases as a function of the tailing coefficient value K1 as shown by the change between the time points t7 and t8, and is set to the maximum retard amount at the time point t8. Then, when the shift to the late lean mode is completed (after time t8 in FIG. 27 (f)), R1 (K1) is set to the retard amount 0. In this way, as the timing of shifting to the late lean mode (time t8 in Fig. 27 (f)) is approached, the retard amount of the ignition timing is increased to adjust the output, and the output sudden change due to the start of the late lean mode control. Is prevented.

After setting the respective combustion parameter values in this way, step S48 is executed and engine control is performed in the first injection mode. Note that step S4 in FIG.
When it is determined that the determination of No. 2 is the lean mode in the previous period (the determination result is affirmative), the ECU 70 determines in steps S43 and S44.
Is executed at the time of transition from the first lean mode to the second lean mode, the details of which will be described later.

When the tailing coefficient value K1 increases and the temporary target air-fuel ratio correction coefficient value Kaft falls below the discrimination value Xaf,
The determination result of step S34 of 7 is affirmative, and the process proceeds to step S36 without executing steps S40, S46, and S47 described above. As a result, the target air-fuel ratio correction coefficient value Kaf is no longer held at the value during S-F / B mode control, but is set to the temporary target air-fuel ratio correction coefficient value Kaft (Kaf = Kaft). As the fuel injection end period Tend and the ignition timing Tig, the values calculated in the late lean mode are used as they are. In this case, FIG.
As can be seen from the change in the correction coefficient value at time t8 in (c),
The target air-fuel ratio correction coefficient value Kaf changes in a step-like manner from a value suitable for that mode corresponding to a value near the theoretical air-fuel ratio in the SF / B mode to a value in the late lean mode that is free from the risk of rich misfire. At this point, the control shifts to the late lean mode. That is, when the engine output during the mode shift control is gradually adjusted by the ignition timing and becomes substantially the same as the output obtained at the rich misfire limit of the air-fuel ratio of about 20 in the late lean mode control, the air-fuel ratio is suddenly changed. Become. Along with this, the injection end period Tend and the ignition timing Tig are also changed to values suitable for the late lean mode control (time t8 in FIGS. 27 (d) and 27 (e)).

After setting each combustion parameter value in this way, step S22 of FIG. 16 is executed, and engine control is performed in the late lean mode. When the tailing coefficient value K1 gradually increases and reaches the value 1.0, it means that the transition to the late lean mode is completed, and the determination result in step S20 of FIG.
s), the engine control is performed in the late lean mode of step S22 after the preparation for the transition to the previous mode control is performed in step S21.
1, S22 will be repeatedly executed.

Next, the transition control from the late lean mode to the early lean mode will be described.
The tailing coefficient K2 in step S6 of FIG.
The value is set to 0 as shown in Fig. 28 (t20 in Fig. 28 (a)).
Time point). In such a case, the ECU 70 executes step S14 after determining the previous mode in step S10 of FIG. Tig, Ten
Computes d, Legr, Ev, etc.

In the lean mode control in the first term, the changeover switches 70a and 70b shown in FIG. 6 are switched to the first term mode by a predetermined tanming described later as in the case of the SF / B mode control described above, and the target average value is set. The effective pressure Pe is calculated according to the intake pipe pressure Pb and the engine speed Ne based on the target average effective pressure map 70r that is the second calculation map. When the target average effective pressure Pe is calculated, the target average effective pressure Pe is calculated using the target air-fuel ratio calculation map 70j, the injection end timing setting means 70m, the ignition timing setting means 70n, E.
It is supplied to the GR amount setting means 70p, respectively, and the target A /
F, Tend, Tig, Legr, Ev, etc. are calculated.

When the calculation of various combustion parameter values and the like is completed as described above, the process proceeds to step S50 of FIG. 19 and it is determined whether or not the tailing coefficient K2 is 1.0. As described above, the tailing coefficient K2 is set to a value 0 immediately after the transition to the lean mode in the previous period, and therefore the determination result in step S50 is No, and the steps after step S51 are executed. Performs the process of shifting from the late lean mode to the early lean mode.

In step S51, it is determined whether or not the invalid period counter Td2 has a value of 0, that is, the initial value f2 (Ne, Pe) of the counter Td2 in the same manner as the transition control from the late lean mode to the SF / B mode. ) Is determined whether or not the invalid period has elapsed. Initial value f2 of counter Td2
(Ne, Pe) is set by the execution of step S21 in FIG. 16 described above, and is equal to the initial value f2 (Ne, Pe) immediately after the mode transition. Therefore, step S5
The determination result of 1 is negative, the process proceeds to step S52, the predetermined value ΔTd2 is subtracted from the counter value Td2, and step S5
In step 3, the tailing coefficient value K2 is reset to the value 0.
Then, these steps S52 and S53 are the invalid period (the period from the time point t20 to the time point t21 shown in FIG. 28 (c)).
Is repeatedly executed until the time elapses, and during that time, the tailing coefficient value K2 is held at the value 0.

Next, the ECU 70 executes steps S55 and S57 to calculate the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency Ev by the above-mentioned formulas (F3) and (F4). During the period in which the coefficient value K2 is 0 (invalid period from time t20 to time t21 shown in FIG. 28), the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency value Ev are the last values, that is, the last lean mode is calculated. When the coefficient value K2 reaches the value 1.0, it is set to the value set by weighting according to the K2 value when the value of the coefficient value K2 increases, and is set to the value calculated by the lean mode in the previous period. It will be. Due to the above-described change in the tailing coefficient value K2, the target air-fuel ratio correction coefficient value Kaf and the volumetric efficiency Ev at the time of mode transition are linearly and gradually increased from time t21 to time t24 as shown in FIG. Is changed, and after t24, it will be held at the value calculated by the lean mode in the previous term.

Next, the ECU 70 proceeds to the above-mentioned step S60 of FIG. 20, and advances the boost pressure delay counter value CNT.
When it is determined that the boost pressure delay counter value CNT2 is not yet reduced to the value 0 (the determination result is negative), step S61 is executed to rewrite the target average effective pressure Pe to the value Pe ', and The value finally calculated in the lean mode is held for a predetermined period (the period from time t20 to time t22 shown in FIG. 28 (b)). The boost pressure delay counter value CNT2 delays the switching of the map for calculating the target average effective pressure Pe. Note that the period corresponding to the initial value XN2 may be set to a value different from the set value in the SF mode described above depending on the engine.

When the determination result of step S60 is affirmative (when the boost pressure delay period has elapsed), the target average effective pressure map 70r shown in FIG. 6 is used for subsequent lean mode control.
The value calculated from is used (t22 in FIG. 27).
After the point). Next, in step S62, the above formula (F3)
It is determined whether or not the temporary target air-fuel ratio correction coefficient value Kaft calculated in step 3 is smaller than the value Xaf. If the target air-fuel ratio correction coefficient value Kaf is smaller than the value Xaf, the temporary target air-fuel ratio correction coefficient value Kaft is calculated.
Until the value reaches the value Xaf (until time t23 shown in FIG. 28 (c)), the target air-fuel ratio correction coefficient value Kaf is changed to the tailing coefficient K2.
Is set to a value corresponding to the above, that is, a temporary target air-fuel ratio correction coefficient value Kaft (step S63). Then, in order to execute the late lean mode control, the ignition timing Tig is held at the final value Tig 'calculated by the late lean mode (step S64), and the injection end period Tend is also the final value calculated by the late lean mode. It is held in Tend '(step S65).

After resetting the respective combustion parameter values in this way, the above-mentioned step S22 of FIG. 16 is executed,
Engine control in the late injection mode is performed. Up to this point, it is the same as the transition control from the late lean mode to the SF / B mode, but when the tailing coefficient value K2 increases and the temporary target air-fuel ratio correction coefficient value Kaft exceeds the determination value Xaf,
Control different from the control for shifting to the S-F / B mode is executed as follows.

That is, in the first-half lean mode, the air-fuel ratio can be set to a leaner side than the stoichiometric air-fuel ratio.
By adjusting the air-fuel ratio, it is possible to control the engine output, and it is possible to prevent a sudden change in the output during transition. Therefore, if the temporary target air-fuel ratio correction coefficient value Kaft exceeds the determination value Xaf and the determination result of step S62 is negative, the process proceeds to step S66, and after it is determined that the lean mode is in the previous period, steps S68 and S6 are performed.
9 is executed, and the target air-fuel ratio correction coefficient value Kaf and the ignition timing Tig are calculated. The target air-fuel ratio correction coefficient value Kaf is rewritten to the temporary target air-fuel ratio correction coefficient value Kaft described above,
The ignition timing Tig is replaced with a value calculated by the following formula (F9). Therefore, the target air-fuel ratio correction coefficient value Kaf is set to a value according to the tailing coefficient value K2 even if the temporary target air-fuel ratio correction coefficient value Kaft exceeds the above-mentioned determination value Xaf.
As shown in 8 (c), it will gradually increase, and when the K2 value reaches the value 1.0, it will shift to the value calculated by the lean mode in the previous term.

Tig = (1−K2) * Tig ′ + K2 * Tig ... (F9) On the other hand, in the shift control to the lean mode in the previous period, S−F / B
There is no setting of the retard amount R2 (K2) used for the shift control to the mode, and the ignition timing Tig is set to a value according to the tailing coefficient value K2. S in the control to shift to lean mode in the previous term
This is because unlike the case of the shift control to the F / B mode, a sudden change in the output can be prevented by adjusting the air-fuel ratio. Therefore, the ignition timing Tig is t2 as shown in FIG. 28 (e).
It suddenly changes at 3 points, and gradually changes toward a value suitable for the previous lean mode control, and at t24 and after that, when the transition to the previous lean mode is completed, the value completely changes to the value calculated in the previous lean mode. .

If the temporary target air-fuel ratio correction coefficient value Kaft exceeds the discriminant value Xaf, step S65 described above is not executed, so that the injection end period Tend uses the value calculated by the lean mode in the previous period as it is. As a result of the above,
In the previous lean mode transition control, the previous lean mode control is executed when the temporary target air-fuel ratio correction coefficient value Kaft exceeds the determination value Xaf, and at that time, the injection end period Tend.
Immediately shifts to the value calculated in the lean mode in the previous period (after time t23 in FIG. 28 (c)), and the ignition timing Tig gradually increases according to the tailing coefficient value K2 (FIG. 2).
(Between t23 and t24 in 8 (e)).

After setting the combustion parameter values in this way, step S48 of FIG. 18 is executed, and the engine control is performed in the first injection mode. Tailing coefficient value K
When 2 gradually increases and reaches the value 1.0, the determination result in step S50 of FIG. 19 becomes affirmative (Yes),
Step S58 is executed. In this determination, it is determined that the lean mode is in the previous period, so the ECU 70 is
Then, the process proceeds to step S80 to prepare for the late lean mode control transition or the S-F / B mode control transition. To prepare for the transition, the initial values of the control variables for the transition are set, and various correction coefficient values K calculated in the current control mode are set.
The af, combustion parameter values Tig, Tend, EV, target average effective pressure Pe, etc. are stored. As the control variables for the transition, there are an invalid period counter Td1 and an EGR delay counter CNT3 described later, and the former counter Td1 is a value set as an initial value according to the target average effective pressure Pe and the engine speed Ne. f1 (Ne, Pe) and the value XN3 are set in the latter counter CNT3, respectively. These transition control variables and the like are updated to new values each time the control in the lean mode is repeated and step S80 is repeatedly executed. Note that this step S80
Then, the EGR delay counter CNT1 used at the time of the transition control from the S-F / B mode to the late lean mode is performed.
The initial value of is not set.

When the initial values of the control variables and the like are set in step S80, the process proceeds to step S82, in which the tailing coefficient value KS used during the transition control from the S / F / B mode to the lean mode is 1.0. Or not.
At present, since the shift to the lean mode has been completed and the control of that mode is being performed, the coefficient value KS is the value 1.0, skipping steps S84 and S86 and proceeding to step S48 of FIG. Control by the injection mode is executed. The jumped steps S84 and S86 are executed during the transition control from the S-F / B mode to the lean mode in the previous period, and the details thereof will be described later.

Next, the shift control when shifting from the first lean mode to the second lean mode will be described. FIG.
When the late lean mode is discriminated during the early lean mode control in step S1 of 4 (at time t25 in FIG. 28 (a)), the tailing coefficient K1 is set to 0 in step S2.
Is set. Then, in step S10 of FIG. 15, the latter mode is discriminated, and in step S12, various combustion parameter values and the like are calculated, and then step S2 of FIG.
0 is executed and it is determined whether K1 is equal to the value 1.0. Immediately after the determination of the late lean mode, the tailing coefficient value K1 is 0, so the determination result of step S20 is negative, and the steps from step S24 are executed to perform the transition process from the early lean mode to the late lean mode. To do.

In step S24, the invalid period counter T
It is determined whether or not d1 is the value 0, that is, whether or not the invalid period corresponding to the initial value f1 (Ne, Pe) of the counter Td1 has elapsed. Initial value f1 of counter Td1 (Ne, P
e) is set by the execution of step S80 in FIG. 22 described above during the first-half lean mode control immediately before the transition, and the counter value Td1 at the time when this step S24 is performed immediately after the mode transition is the initial value f1. (Ne, P
equal to e). Therefore, the determination result in step S24 is negative, the process proceeds to step S25, the predetermined value ΔTd1 is subtracted from the counter value Td1, and the tailing coefficient value K1 is reset to 0 in step S26. Then, these steps S25 and S26 are repeatedly executed until the invalid period elapses (until the time t26 in FIG. 28C), and the tailing coefficient value K1 is held at the value 0 during that period.

Next, the ECU 70 executes step S28 and step S30 of FIG. 17 to set the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency Ev to the above-mentioned S-F.
Similar to the transition control from the / B mode to the late lean mode, the above equations (F6) and (F7) are used for the calculation. in this case,
Kaf 'and Ev' in the equations (F6) and (F7) are the target air-fuel ratio correction coefficient value and the volume efficiency calculated last in the lean mode, respectively, and the step S80 in FIG. It is stored as Kaf 'value and Ev' value at the time of doing.

Therefore, the period in which the coefficient value K1 is 0 (invalid period from time t25 to time t26 shown in FIGS. 28 (c) and 28 (f))
Is the temporary target air-fuel ratio correction coefficient value Kaft and the volumetric efficiency value E
v is held at the previous value, that is, the value calculated last by the lean mode in the first half, but when the value of the coefficient value K1 increases, the coefficient value K1 is set to the value set by weighting according to the K1 value.
When reaches the value of 1.0, they will be set to the values set during the late lean mode control. Due to the change in the tailing coefficient value K1 as described above, the target air-fuel ratio correction coefficient value Kaf and the volumetric efficiency Ev at the time of mode transition are shown in FIG.
The value is gradually changed linearly as shown from the time t26 to the time t28 in 8 (c) and (f), and after t28, it is held at the value set by the late lean mode control. .

Next, the ECU 70 executes step S in FIG.
In step 31, it is determined whether or not the EGR delay counter CNT1 has counted down to the value 0. This counter CN
T1 is provided for the purpose of delaying the EGR control in the late lean mode. However, during the transition control from the early lean mode to the late lean mode, T1 shown in FIG.
As described in step S80, the counter CN
The initial value of T1 is not set (CNT1 = 0). That is, during the transition control of this time, the EGR control is not delayed, and the valve opening degree Leg of the EGR valve 45 is determined at step S31 at the time when the mode transition is determined (at time t25 in FIG. 28 (a)). The result is affirmative, and the value is immediately set to the value calculated by the late lean mode (the value calculated by the EGR amount setting means 70p in FIG. 6) without executing step S32, and the valve opening degree of the EGR valve 45 is set to this value. Will be controlled (after time t25 in FIG. 28 (g)). When shifting from the early lean mode to the late lean mode, the EGR amount is immediately shifted to the calculated value by the late lean mode control at the same time when the shift determination is made, because the difference in the magnitude of the EGE amount calculated between the two modes is large. Even if the EGR valve 45 is opened at the same time as the transition determination, the EGE is actually
Because there is a time lag for the amount to change.

Then, the process proceeds to step S34, and the previous equation (F6)
It is determined whether or not the temporary target air-fuel ratio correction coefficient value Kaft calculated in step 1 is smaller than the value Xaf, and it is determined whether or not the late lean mode may be started. If the determination result of step S34 is negative and the target air-fuel ratio correction coefficient value Kaf is greater than or equal to the value Xaf, the previous lean mode control is continuously executed. That is, the ECU 70 determines that the determination result of step S34 is negative, that is, until the temporary target air-fuel ratio correction coefficient value Kaft reaches the value Xaf (from time t26 shown in FIG. 28 (c) to t2.
18), the process proceeds to step S40 in FIG. 18, rewrites the injection end period Tend to the last value Tend 'calculated in the lean mode in the previous period, and holds this value (time t26 to time t27 shown in FIG. 28 (d)). Until)). If the control mode before the transition determination is the lean mode in the previous period, S
In order to determine whether it is the -F / B mode, it is determined whether or not the correction correction coefficient value Kaf 'set and stored immediately before the transition determination is smaller than the value 1.0.

Since the shift control from the lean mode in the previous period is being performed, the result of this determination in step S42 is affirmative,
In step S43, the target air-fuel ratio correction coefficient value Kaf is rewritten to the temporary target air-fuel ratio correction coefficient value Kaft. The ignition timing Tig is calculated by the following equation (F
Replace with the value calculated by 10). As a result, the target air-fuel ratio correction coefficient value Kaf and the ignition timing Tig are
As shown from t26 to t27 in (c) and (e),
It gradually changes according to the tailing coefficient value K1.

Tig = (1−K1) * Tig ′ + K1 * Tig ... Although the sudden change of output was prevented, the retard amount R1 was added to the above formula (F10).
(K1) is not included. In the case of the transition from the early lean mode to the late lean mode, the output control is performed by adjusting the air-fuel ratio, and therefore the correction by the retard amount R1 (K1) is not necessary and the ignition timing Tig is the tailing coefficient value K1. It is set to a value according to.

After setting each combustion parameter value in this way, step S48 is executed, and engine control is performed in the first injection mode. When the tailing coefficient value K1 increases and the provisional target air-fuel ratio correction coefficient value Kaft falls below the determination value Xaf, the determination result of step S34 in FIG. 17 becomes affirmative, and the steps S40 and S44 described above are not executed. It proceeds to step S36. As a result, the target air-fuel ratio correction coefficient value Kaf is set to the temporary target air-fuel ratio correction coefficient value Kaft (Kaf = Kaft), and accordingly, the fuel injection end period Tend and the ignition timing Tig are calculated in the late lean mode. The value is used as is. In this case, the injection end period Tend and the ignition timing Tig are stepwise changed to values suitable for the late lean mode control, as can be seen from the changes at time t27 in FIGS. 28 (d) and 28 (e).

After setting the combustion parameter values in this way, step S22 of FIG. 16 is executed, and the engine control is performed in the late injection mode. Tailing coefficient value K
When 1 gradually increases and reaches the value of 1.0, it means that the transition to the late lean mode has been completed, and as shown in FIG.
The determination result in step S20 of 6 is positive (Yes)
In step S21, the engine control is performed in the late injection mode in step S22 after the preparation for the transition to the early mode control is performed.
1, S22 are repeatedly executed.

Next, the transition control for shifting from the lean mode to the SF / B mode in the first half will be described. The tailing coefficient K in step S8 of FIG.
L is set to the value 0 as shown in FIG.
9 at t30). In such a case, the ECU 70 is
After determining the previous mode in step S10 of 5,
In step S14, the various combustion parameter values Pe, Kaf, Tig, Tend, Legr, Ev, etc. described above are calculated.

When the calculation of various combustion parameter values etc. is completed, the routine proceeds to step S50 of FIG. 19 and the tailing coefficient K
Determine if 2 is the value 1.0. This tailing coefficient value K2 is the SF from the lean mode in the previous term to the SF.
Since it is a loop immediately after the transition to the / B mode, the value should be 1.0, and the determination result of step S50 is affirmative, and the process proceeds to step S58. Next, in step S58, after it is determined that the S-F / B mode is set, the ECU 70
21 proceeds to step S70 in FIG. 21 and prepares for the above-described late lean mode control transition or early lean mode control transition. Then, after completing the preparation for the transition, the process proceeds to step S72.

Immediately after the transition to the SF / B mode is determined, the tailing coefficient value KL has just been set to the value 0, so the determination result in step S72 is negative, and the volume efficiency in step S73 is negative. Ev is the following equation
Calculate based on (F11). Ev = (1−KL) * Ev ′ + KL * Ev ... (F11) The above formula (F11) is similar to the above formula (F4), and Ev ′
Is the volumetric efficiency calculated last in the lean mode in the previous period, and is stored as the Ev ′ value when the step S80 of FIG. Ev of the last term on the right side of the above equation is a value calculated in the S-F / B mode this time.

Therefore, the volumetric efficiency value Ev is calculated by the S-F / B mode when the coefficient value KL reaches a value of 1.0 when the coefficient value KL increases to a value set by weighting according to the KL value. It will be set to each value. Due to the change in the tailing coefficient value KL as described above,
The volumetric efficiency Ev at the time of mode transition is linearly gradually changed as shown from the time t30 to the time t32 in FIG. 29 (f), and is calculated by the SF mode after the time t32. Will be retained in the value.

Next, the ECU 70 executes the step S in FIG.
Proceeding to 74, it is determined whether or not the EGR delay counter CNT3 has counted down to the value 0. This counter CN
T3 is provided for the purpose of delaying the EGR control in the S-F / B mode, which stabilizes the control during the mode transition. The counter value CNT3
When the lean mode control is repeatedly executed in the previous period, the initial value XN3 is reset each time in step S80 of FIG. 22 described above, and the crank angle sensor 17
Each time when detects a predetermined crank angle position, the crank interrupt routine shown in FIG. 25 is executed and the countdown is performed.

If the determination result in step S74 is negative, that is, if the EGR delay period (the period corresponding to the initial value XN3 shown from the time t30 to the time t31 in FIG. 29 (g)) has not elapsed, the EGR valve 45 The valve opening Legr is the previous value,
That is, it is set to the value Legr 'during the previous lean mode control executed immediately before the determination of the shift to the S-F / B mode.
This value Legr 'is stored and updated each time the above-described step S80 of FIG. 22 is executed. When the determination result of step S74 is affirmative, that is, when the period corresponding to the initial value XN3 has elapsed (after time t31 in FIG. 29 (g)), the valve opening L
egr is set to a value calculated in the SF / B mode, and the valve opening degree of the EGR valve 45 is controlled based on the set valve opening degree Legr.

The transition from the lean period mode to the SF / B mode is the transition in the same early injection mode.
The same map is used as the map for calculating the target average effective pressure Pe (see FIG. 29 (b)). Further, since the driver's output request is assumed by depressing the accelerator pedal, the target air-fuel ratio correction coefficient value Kaf, the fuel injection end period Tend, and the ignition timing Tig are immediately determined at time t30 when the mode transition is determined. The values are switched to the respective values calculated in the S-F / B mode (see FIGS. 29 (c), (d) and (e)).

After calculating the various combustion parameter values in this way, the process proceeds to the step S48 of FIG. 18, and the control in the first-term injection mode is executed. Finally, S-F / B
Transition control when the mode shifts to the lean mode in the first half will be described. In this case, the ECU 70 determines the tailing coefficient value KS by executing step S6 of FIG.
The value is set to 0 according to the control rule shown in FIG. Then, after determining the previous mode in step S10 of FIG. 15, various combustion parameter values and the like are calculated in step S14, and the process proceeds to step S50 of FIG. In step S50, it is determined whether or not the tailing coefficient value K2 is 1.0, but since the current loop is immediately after the transition from the SF / B mode to the lean mode is determined, the coefficient value K2 is equal to the value. Since it should be 1.0, the determination result is affirmative and the process proceeds to step S58. In this step, the previous lean mode is determined, and step S80 of FIG. 22 is executed. In step S80, as described above, in order to prepare for the late lean mode control transition or the S-F / B mode control transition, the initial values of the control variables for the transition are set, and various corrections calculated in the current control mode are performed. Coefficient value Kaf, combustion parameter values Tig, Tend, EV, target average effective pressure Pe
Etc. are memorized.

When the initial values of the control variables and the like are set in step S80, the process proceeds to step S82, in which the tailing coefficient value KS used during the transition control from the SF / B mode to the lean mode is 1.0. Or not.
The present loop is immediately after the determination of the shift to the lean mode in the previous term, and the tailing coefficient value KS has just been set to the value 0, so the determination result of step S82 is negative. When the determination result of step S82 is negative, the ECU 70 repeatedly executes step S84 and step S86, and in step S84, calculates the volumetric efficiency Ev based on the following formula (F12).

Ev = (1−KS) * Ev ′ + KS * Ev ... (F12) The above formula (F12) is similar to the above formulas (F11) and (F4), and Ev ′ is It is the volume efficiency finally calculated in the S-F / B mode, and is stored as the Ev ′ value when the above-described step S70 of FIG. 21 is finally executed. Ev of the last term on the right side of the above equation is a value calculated in the lean mode in the previous term this time.

Therefore, the volumetric efficiency value Ev is a value calculated by weighting according to the KS value when the coefficient value KS increases, and a value calculated by the lean mode in the first half when the coefficient value KS reaches the value 1.0. Will be set respectively. Due to the change in the tailing coefficient value KS as described above,
The volume efficiency Ev at the time of mode transition is the tailing coefficient value KS
Over the period in which the value changes from the value 0 to the value 1.0,
The value is gradually changed linearly as shown from the time point t34 to the time point t35, and after the time point t35, it is held at the value calculated by the lean mode in the previous period.

Next, at step S86, the target air-fuel ratio correction coefficient value Kaf, the ignition timing Tig, and the injection end period Tend.
Is the last calculated value K in S-F / B mode
It is set to af ', value Tig', and value Tend ', and these values are held until the tailing coefficient value KS reaches the value 1.0 (at time t34 in FIGS. 29 (c), (d), and (e)). From t35 to t35).

Kaf = Kaf 'Tig = Tig' Tend = Tend 'As described above, the tailing coefficient value KS changes from the value 0 to the value 1.0 at the time of the transition from the SF / B mode to the lean mode.
Until the tailing coefficient value KS reaches 1.0, the target air-fuel ratio correction coefficient value Kaf, the ignition timing Tig, and the fuel injection end timing Tend. Is switched to the previous lean mode, at which point the transition to the previous lean mode control is completed. In the transition control from the S-F / B mode to the lean mode in the previous period, these combustion parameter values may be gradually changed according to the tailing coefficient value from the viewpoint of prevention of switching shock, but they are gradually changed. Since the exhaust gas characteristics (particularly the NOx emission amount) may be deteriorated at the time of switching, in the present invention, switching shock is prevented by gradually changing the volumetric efficiency Ev, and the volumetric efficiency Ev conforms to the lean mode in the previous period. The target air-fuel ratio correction coefficient value Kaf, the ignition timing Tig, and the fuel injection end timing Tend at a time when the tailing coefficient value KS reaches a value of 1.0 (the tailing coefficient value KS reaches a value of 1.0) at a stroke. From the previous term to a value compatible with the lean mode to minimize the generation of NOx and the like.

During the transition control from the SF / B mode to the lean mode in the previous period, the valve opening degree Leg of the EGR valve 45 is set to a value calculated in the lean mode in the previous period at the same time when the mode transition is determined ( Refer to time point t34 in FIG. 29 (g)). S-F / that controls the air-fuel ratio near the stoichiometric air-fuel ratio
In the B mode control, the three-way catalyst 42 shown in FIG. 1 suppresses the emission of nitrogen oxides NOx, but when the air-fuel ratio is leaner than the stoichiometric air-fuel ratio, the lean air-fuel ratio (for example, the air-fuel ratio 22) is used. It is better to introduce a large amount of exhaust gas into the engine 1 at an early stage. Therefore, the valve opening degree Legr of the EGR valve 45 is changed at the same time as the mode transition determination.

When the setting of the various combustion parameter values and the like is completed in this way, the ECU 70 proceeds to step S48 of FIG.
Then, the engine control in the first term injection mode is executed. In the above-described embodiment, as is clear from the comparison of the timing charts shown in FIGS. 27 to 29, when switching the map for calculating the target average effective pressure Pe, when switching from the late lean mode to the early injection mode, An invalid period (a period corresponding to XN2) is provided, and the intake pipe pressure Pb
Was switched after waiting for the response delay of
When the mode is switched from the F / B mode or the previous lean mode) to the latter lean mode, the switching is immediately performed when the mode transition is determined. During the late lean mode control, the EGR valve 45 and the ABV 27 are opened to supply a large amount of exhaust gas and bypass air to the engine 1, and the overall air-fuel ratio is set to an extremely large value (for example, 30 to 35). Since the output control of the engine 1 is performed by adjusting the fuel injection amount regardless of the EGR amount and the fresh air intake air amount, unlike the parameters such as the intake pipe pressure Pb with a detection delay, the throttle valve 28 By detecting the valve opening degree θth, the driving intention of the driver is reflected in the control without delay, and the engine control with good responsiveness can be performed.

[0129]

In the control apparatus for a direct injection spark ignition internal combustion engine according to claim 1 of the present invention, the target average effective pressure information is calculated based on at least the engine speed and the load information, and the calculated target average effective pressure information is calculated. Parameter values that affect the combustion state in the combustion chamber, such as the fuel injection amount, the ignition timing, and the exhaust gas recirculation amount (claim 2), are set according to the target average effective pressure information. If the combustion parameter value is associated in advance, even if the passage area of the throttle body is changed, for example, it is only necessary to associate the relationship between the throttle valve opening and the average effective pressure information. There is an effect that it is not necessary to spend time and labor every time the engine design is changed to reset the map or the like for calculating the combustion parameter value.

Further, in the control device of the third aspect, there is provided a switching means for switching the control of the engine between the early injection mode and the late injection mode, and the target average effective pressure information can be switched also in such an engine. By calculating according to the mode, the combustion parameter setting means only needs to set the combustion parameter value according to the calculated target average effective pressure information. It can be performed.

According to the control device of the fourth aspect, the target average effective pressure information set by the calculating means uses the first and second calculation maps prepared in advance for each mode, so that the computer control is possible. Become suitable. According to the control device of claim 5, the late injection mode is switched at the time of low load operation such as idle operation, and from the second calculation map used in the late injection mode, the throttle valve indicating the driver's driving intention is displayed. Since the target average effective pressure information is calculated according to the opening degree and the engine speed, according to the control device of claim 6, when the early injection mode is switched to the late injection mode, Since the target average effective pressure information is calculated using the second calculation map at the same time as the mode switching, the engine control with extremely excellent responsiveness is performed in combination with the direct injection of fuel into the cylinder. be able to.

On the other hand, according to the control device of the seventh to tenth aspects, when the throttle valve is normally opened, the operation is switched to the first term injection mode, and from the first calculation map used in the first term injection mode, The target average effective pressure information is calculated according to the pressure in the intake passage (Claim 7), the fresh air intake air amount (Claim 8) or the volumetric efficiency (Claims 9 and 10), and the engine speed. Similar to the intake pipe injection type engine, it is possible to perform highly accurate engine control by using the intake air amount information having a high correlation with the output characteristic.

According to the control device of the eleventh aspect, when the throttle valve is opened and the late injection mode is switched to the early injection mode, the first calculation map is used after waiting a predetermined period from the time of mode switching. Since the target average effective pressure information is calculated in this manner, the engine control error caused by the response delay until the intake passage pressure and the fresh air intake air amount reach the value corresponding to the throttle valve opening is avoided. be able to.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing an embodiment of an engine control device according to the present invention.

FIG. 2 is a vertical sectional view of a cylinder injection gasoline engine according to an embodiment.

FIG. 3 is related to an embodiment, which is defined according to an engine in-cylinder average effective pressure Pe and an engine speed Ne, and shows a late injection lean operation range, an early injection lean operation range, an early injection stoichio feedback operation range, and the like. It is a fuel injection control map.

FIG. 4 is an explanatory diagram showing a fuel injection mode in a late injection mode in the embodiment.

FIG. 5 is an explanatory diagram showing a fuel injection mode in a first-term injection mode in the embodiment.

FIG. 6 Target average effective pressure Pe, target air-fuel ratio correction coefficient value K
af, fuel injection end period Tend, basic ignition timing θB, EG
FIG. 6 is a block diagram showing a procedure for calculating a valve opening degree Legr of the R valve 45 and the like.

7 shows a schematic configuration of a target average effective pressure calculation map 70c of FIG. 6, and a target average effective pressure P calculated according to a valve opening degree θth of a throttle valve 28 and an engine speed Ne.
It is a figure for demonstrating e.

FIG. 8 is a diagram showing a schematic configuration of a target average effective pressure calculation map 70r of FIG. 6, and is a diagram for explaining a target average effective pressure Pe calculated according to an intake pipe pressure Pb and an engine speed Ne.

FIG. 9 is used during the late lean mode control, and the volumetric efficiency E is determined according to the target average effective pressure Pe and the engine speed Ne.
It is a figure which shows the structure of the map for calculating v.

FIG. 10 is a diagram showing a configuration of a map which is used during the first-term injection mode control and is used to calculate a volumetric efficiency Ev according to an intake pipe pressure Pb and an engine speed Ne.

11 is a target air-fuel ratio correction coefficient value calculation map 70 of FIG.
j shows a schematic configuration of j, and a target air-fuel ratio correction coefficient Kaf calculated according to the target average effective pressure Pe and the engine speed Ne.
It is a figure for explaining.

FIG. 12 is a diagram showing a schematic configuration of an ignition timing setting means 70n of FIG. 6, and is a diagram for explaining a basic ignition timing θig calculated according to a target average effective pressure Pe and an engine speed Ne.

13 is a diagram showing a schematic configuration of EGR setting means 70p of FIG. 6 and is a diagram for explaining a valve opening degree Leg of an EGR valve 45 calculated according to a target average effective pressure Pe and an engine speed Ne. .

FIG. 14 is a part of a flowchart of a combustion parameter setting routine for setting various combustion parameter values.

15 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

16 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

FIG. 17 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

FIG. 18 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

19 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

20 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

21 is another part of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

22 is the rest of the flowchart of the combustion parameter setting routine following the flowchart of FIG.

FIG. 23 shows E every time a clock pulse of a predetermined cycle is generated.
9 is a part of a flowchart of a timer routine executed by the CU 70.

24 is the rest of the flowchart of the timer routine following the flowchart of FIG. 23.

FIG. 25 is a flowchart of a crank interrupt routine executed by the ECU 70 each time a predetermined crank angle position of the engine 1 is detected.

FIG. 26 is a diagram for explaining various tailing coefficient values that are used during mode transition control and are set according to the control mode transition mode.

FIG. 27 is a timing chart showing changes over time of various control variables and combustion parameter values during mode transition control between the late lean mode and the SF / B mode.

FIG. 28 is a timing chart showing changes over time of various control variables and combustion parameter values during mode transition control between the late lean mode and the early lean mode.

FIG. 29 is a timing chart showing changes with time of various control variables and combustion parameter values during mode transition control between the lean mode and the SF / B mode in the first period.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 engine 4 fuel injection valve 5 combustion chamber 17 crank angle sensor (operating state detecting means) 24 ISCV 25 intake pipe 26 air bypass pipe 27 ABV 28 throttle valve 29 throttle sensor (operating state detecting means) 31 boost pressure sensor (operating state detection) 70 ECU 70a Changeover switch (changeover means) 70b Changeover switch (changeover means) 70c Target average effective pressure map (calculation means / first calculation map) 70j Target air-fuel ratio correction coefficient value calculation map (combustion parameter setting means) 70m Injection End timing setting means (combustion parameter setting means) 70n Ignition timing setting means (combustion parameter setting means) 70p EGR amount setting means (combustion parameter setting means) 70r Target average effective pressure map (calculation means / second calculation map)

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication location F02D 45/00 364 F02D 45/00 364Z (72) Inventor Hiromitsu Ando Shiba 5-chome 33, Minato-ku, Tokyo No. 8 Mitsubishi Motors Corporation

Claims (11)

[Claims] 1. A control device for a cylinder injection type spark ignition internal combustion engine for directly injecting fuel into a combustion chamber, comprising: operating state detecting means for detecting at least engine speed and load information; and detecting by the operating state detecting means. Calculating means for calculating target average effective pressure information based on the result, and combustion parameter setting means for setting a parameter value affecting the combustion state in the combustion chamber according to the target average effective pressure information calculated by the calculating means. And a control unit for controlling the internal combustion engine based on a parameter value set by the combustion parameter setting means. 2. The parameter value that affects the combustion state is at least one of a fuel injection amount, a fuel injection end timing, an ignition timing, and an exhaust gas recirculation amount that recirculates exhaust gas to an intake side. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 1. 3. A switching means for switching between a first-stage injection mode in which fuel injection is mainly performed in an intake stroke and a second-stage injection mode in which fuel injection is mainly performed in a compression stroke, according to a detection result of the operating state detection means. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 1 or 2, wherein the calculation means calculates the target average effective pressure information in accordance with the mode switched by the switching means. 4. The calculation means is used when switching to the first-stage injection mode, and is used when switching to the second-stage injection mode and a first calculation map for calculating the target average effective pressure information. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 3, further comprising: a second calculation map for calculating the target average effective pressure information. 5. The load information detected by the operating state detecting means is a throttle valve opening, and the calculating means calculates the target average from the second calculating map according to at least the engine speed and the throttle valve opening. The control device for a cylinder injection type spark ignition type internal combustion engine according to claim 4, wherein the effective pressure information is calculated. 6. The calculation means calculates the target average effective pressure information by using the second calculation map at the same time when the switching is switched from the early injection mode to the late injection mode by the switching means. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 5, wherein 7. The load information detected by the operating state detecting means is an intake passage internal pressure downstream of the throttle valve, and the calculating means is based on at least the engine speed and the intake passage internal pressure. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 4, wherein the target average effective pressure information is calculated from: 8. The load information detected by the operating state detecting means is a fresh air intake air amount, and the calculating means calculates the target from the first calculation map according to at least the engine speed and the fresh air intake air amount. The control device for a cylinder injection type spark ignition internal combustion engine according to claim 4, wherein average effective pressure information is calculated. 9. The in-cylinder injection type according to claim 4, wherein the calculation means calculates the target average effective pressure information from the first calculation map according to at least the engine speed and the volumetric efficiency. Control device for spark ignition internal combustion engine. 10. The in-cylinder injection spark according to claim 9, wherein the volumetric efficiency is calculated based on either one of the fresh air intake air amount and the intake passage internal pressure downstream of the throttle valve. Ignition type internal combustion engine control device. 11. When the switching means switches from the late injection mode to the early injection mode, the calculating means calculates the first time after a lapse of a predetermined period from the time of switching.
The control device for a cylinder injection type spark ignition internal combustion engine according to any one of claims 4 to 10, wherein the target average effective pressure information is calculated using a calculation map.