JPH10231745A - Fuel injection controller for intercylinder direct injection engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attempt compatibility between improvement of fuel consumption and improvement of exhaust emission by preventing smoking, improving ignitability, and improving the lean limit of an air-fuel ratio. SOLUTION: When stratified burning is selected, fuel injection is executed twice per one cycle of one air cylinder, and as a first time fuel injection, an igniting fuel injection amount GFFST by which at least an air-fuel mixture which can be ignited by an ignition plug can be formed is set (S34), and time or a crank angle that the air-fuel mixture by injection fuel reaches the ignition plug after injection fuel is collided with the top surface of a piston after the fuel of this igniting fuel injection amount is ejected, is counted backward on the basis of ignition timing TADV, then this time or crank angle before ignition is set as igniting fuel injection starting timing IJSTFS (S40). Also, as a second time fuel injection, a main fuel injection amount GFMAIN adapted for stratified burning is set on the basis of an engine speed NE and a throttle opening α(S35), and main fuel injection starting timing IJSTMAIN to eject fuel by the main fuel injection amount adapted for to the stratified burning, is set (S41).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection control device for a direct injection engine in a cylinder, which injects fuel directly into a cylinder and ignites the injected fuel by a spark plug and burns the fuel. The present invention relates to a fuel injection control device for a direct injection engine that can achieve both improvement of fuel efficiency and improvement of exhaust emission by improving an air-fuel ratio lean limit.

[0002]

2. Description of the Related Art Conventionally, in a cylinder direct injection engine in which fuel is directly injected into a cylinder and the injected fuel is ignited by a spark plug and burned, exhaust emissions, particularly nitrogen oxides (hereinafter abbreviated as "NOx"). In order to reduce), if the spray of the injected fuel and the air in the cylinder are not mixed so much that the fuel droplets are present in the initial stage of the combustion, the combustion temperature decreases and the generation of NOx decreases. It is known.

In order to improve the ignitability of the in-cylinder direct injection engine, a flammable air-fuel mixture must be present near the discharge electrode of the ignition plug at the time of ignition. To cope with this, when the fuel injected by the injector (fuel injection valve) is directed to the discharge electrode of the ignition plug, fuel droplets adhere to the discharge electrode of the ignition plug and carbon deposits on the discharge electrode, thereby causing smoldering. There are inconveniences.

For this reason, as disclosed in Japanese Patent Application Laid-Open No. 4-183922, the fuel injection direction of the injector is arranged without directing to the discharge electrode of the ignition plug, and the fuel spray from the injector is sprayed to the top of the piston. The air-fuel ratio in the flammable range that rebounds is directed to the discharge electrode of the ignition plug by a synergistic effect with the in-cylinder intake air flow, and the mixture is ignited by the ignition plug. Therefore, a technique for improving ignitability while preventing smoldering is known.

[0005]

SUMMARY OF THE INVENTION However, NOx
When setting the fuel injection timing for the purpose of reducing the occurrence of
In order for fuel droplets to be present in the early stage of combustion without mixing the spray of injected fuel and air in the cylinder so much, it is necessary to inject fuel at a timing immediately before ignition (ignition timing). In addition, when setting the fuel injection timing for the purpose of improving the ignitability, it is necessary to allow time for the fuel spray from the injector injected fuel to collide with the piston top surface and reach the rebound spark plug. It is necessary to inject the fuel much earlier.

Accordingly, the fuel injection timing for the purpose of reducing the generation of NOx and the fuel injection timing for the purpose of improving the ignitability are opposite to each other. Both cannot be satisfied by one fuel injection per cylinder cycle.

[0007] Japanese Patent Application Laid-Open No. 2-169834 discloses that
The required fuel injection amount according to the engine operating state can be divided into an intake stroke and a compression stroke, and the fuel injection is performed twice per one cylinder cycle, and the first fuel injection is performed in the intake stroke. Forming a uniform lean mixture,
A technique for improving the ignitability and combustibility by enabling the formation of a stratified mixture that can be ignited and burned by forming a rich mixture near the spark plug by performing the second fuel injection in the compression stroke Is disclosed.

However, in the above-mentioned prior art, a rich fuel-air mixture is formed near the spark plug in the second fuel injection in the compression stroke, so that the fuel injection direction from the injector must be directed to the discharge electrode of the spark plug. Yes (see FIG. 3 of the publication), as described above, there is a disadvantage that fuel droplets adhere to the discharge electrode of the ignition plug and carbon deposits on the discharge electrode, resulting in smoldering. Further, it is necessary to diffuse the lean mixture in the combustible range into the cylinder in the first fuel injection in the intake stroke, and the air-fuel ratio lean limit is low, so that the fuel efficiency cannot be sufficiently improved.

The present invention has been made in view of the above circumstances, and has been made in consideration of the above circumstances, a direct injection engine in a cylinder capable of preventing smoldering and achieving both improved ignitability and improved exhaust emission by improving an air-fuel ratio lean limit. It is an object of the present invention to provide a fuel injection control device.

[0010]

In order to achieve the above object, the invention according to claim 1 directly injects fuel into a cylinder,
In a fuel injection control device for a direct injection engine in which an injected fuel is ignited by a spark plug and burns, as shown in the basic configuration diagram of FIG. Ignition fuel injection amount setting means for setting the ignition fuel injection amount to be obtained based on the operation state; main fuel injection amount setting means for setting the main fuel injection amount for obtaining a predetermined output according to the operation state; The time or crank angle from when the fuel is injected for the fuel injection amount to when the mixture of the injected fuel reaches the spark plug after the injected fuel collides with the piston top surface is calculated back with reference to the ignition timing. Ignition fuel injection timing setting means for setting a time before ignition or a crank angle as an ignition fuel injection timing; and a main fuel injection unit for injecting fuel according to the main fuel injection amount after the ignition fuel injection. Characterized by comprising a main fuel injection timing setting means for setting on the basis of fuel injection timing in the operation state.

According to a second aspect of the present invention, fuel is injected directly into a cylinder to ignite the injected fuel by a spark plug and burn, and selectively performs stratified combustion and uniform mixed combustion in accordance with the operating state of the engine. In the fuel injection control device for the engine to be performed, as shown in the basic configuration diagram of FIG. 1B, stratified charge combustion is selected at low rotation and low load, and uniform mixed combustion is selected at high rotation and high load, based on the engine operating state. Combustion mode selection means; ignition fuel injection quantity setting means for setting an ignition fuel injection quantity capable of at least forming an air-fuel mixture ignitable by a spark plug when stratified combustion is selected; stratified combustion according to an engine operating state A main fuel injection amount setting means for setting a main fuel injection amount adapted to the fuel injection amount, and a fuel injection amount for the ignition fuel injection amount, and after the injected fuel collides with the piston top surface, the mixed fuel is used. Ignition fuel injection timing setting means for calculating the time or crank angle until the gas reaches the ignition plug with reference to the ignition timing and setting the time or crank angle before ignition as the ignition fuel injection timing; Main fuel injection timing setting means for setting a main fuel injection timing for injecting fuel according to the main fuel injection amount adapted to combustion based on an engine operating state; and, when uniform mixed combustion is selected, uniform mixed combustion based on the engine operating state. Fuel injection amount setting means for setting a fuel injection amount adapted to the fuel injection amount, and fuel injection timing setting means for setting a fuel injection timing for injecting fuel based on the fuel injection amount adapted to the uniform mixed combustion based on an engine operating state. It is characterized by having.

That is, according to the first aspect of the present invention, an ignition fuel injection amount capable of forming an air-fuel mixture ignitable by a spark plug at a minimum is set based on the operating state, and a predetermined output is output according to the operating state. Set the main fuel injection amount to obtain. The time or crank angle from when the fuel is injected at the ignition fuel injection amount to when the mixture of the injected fuel reaches the spark plug after the injected fuel collides with the piston top surface, based on the ignition timing. The time before the ignition or the crank angle is set as the ignition fuel injection timing. Further, a main fuel injection timing for injecting fuel based on the main fuel injection amount after the ignition fuel injection is set based on the operating state.

According to the second aspect of the present invention, stratified charge combustion is selected at low rotation speed and low load, and uniform mixed combustion is selected at high rotation speed and high load based on the operating state of the engine. When stratified charge combustion is selected, an ignition fuel injection amount that can minimize the mixture ignitable by the spark plug is set, and a main fuel injection amount adapted to stratified charge combustion is set according to the engine operating state. . The time or crank angle from when the fuel is injected at the ignition fuel injection amount to when the mixture of the injected fuel reaches the spark plug after the injected fuel collides with the piston top surface, based on the ignition timing. The time before the ignition or the crank angle is set as the ignition fuel injection timing. Further, the main fuel injection timing for injecting the fuel with the main fuel injection amount adapted to the stratified combustion is set based on the engine operating state. On the other hand, when uniform mixed combustion is selected, a fuel injection amount adapted to uniform mixed combustion is set based on the engine operating state. Then, the fuel injection timing for injecting the fuel with the above fuel injection amount adapted to the uniform mixed combustion is set according to the engine operating state.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

First, the schematic configuration of the in-cylinder direct injection engine will be described with reference to FIG. Reference numeral 1 denotes a horizontally opposed 4-cycle 4-cylinder in-cylinder direct injection gasoline engine (hereinafter simply referred to as "engine") for a vehicle such as an automobile as an example of an in-cylinder direct injection engine. Cylinder heads 2 are provided on both left and right banks of a cylinder block 1a of the engine 1, respectively.
And an exhaust port 2b.

In the intake system of the engine 1, an intake manifold 3 communicates with each intake port 2a, and a throttle chamber 5 communicates with the intake manifold 3 via an air chamber 4 in which intake passages of respective cylinders gather. An air cleaner 7 is attached to the upstream side of the throttle chamber 5 via an intake pipe 6, and the air cleaner 7 is connected to the air intake chamber 8.

The throttle chamber 5 is provided with a throttle valve 5a linked to an accelerator pedal. The intake pipe 6 is connected to a bypass passage 9 that bypasses the throttle valve 5a.
At the time of idling, the bypass passage 9 depends on the valve opening.
An idle speed control valve (ISC valve) 10 for controlling the idle speed by adjusting the amount of bypass air flowing through the engine is provided.

On the other hand, as shown in FIGS. 16 and 17, the combustion chamber structure of the engine 1 has a piston 1 for each cylinder.
A combustion chamber 12 is formed by the top surface 11a of the cylinder head 1, the inner wall of the cylinder block 1a, and the inner recess 2c formed on the bottom surface of the cylinder head 2.

The inner recess 2 has a pent roof shape, and the top 2d is slightly deviated from the center of the cylinder bore toward the exhaust port 2b. An injector 13 is provided for each cylinder substantially at the center of the top 2d of the pent roof inner surface concave portion 2c, and an injection hole 13a of the injector 13 is directed horizontally to the combustion chamber 12. In this embodiment, the engine is a four-valve engine. Intake ports 2a are opened on both sides of the injector 13 of the intake side pent roof surface 2e of the inner recess 2c of the cylinder head 2, and the exhaust port pent roof surface 2f is opened. Injector 13
The exhaust ports 2b are respectively opened on both sides sandwiching.
A squish area 12a is formed at the bottom of each of the pent roof surfaces 2e and 2f of the inner recess 2c of the cylinder head 2.

The intake port 2a and the exhaust port 2b are opened and closed at a predetermined timing by an intake valve 16 and an exhaust valve 17 linked to the intake cam 14 and the exhaust cam 15. The intake port 2a is formed in a straight port shape as shown in FIG. 16, and when the intake valve 16 is opened, the intake air causes the in-cylinder intake air to flow from the intake port 2a to the exhaust side pent roof surface 2f. In the intake air flowing into the combustion chamber 12 and flowing into the combustion chamber 12, a so-called tumble flow is generated in the vertical direction with respect to the combustion chamber 12, and the tumble flow enables stratified combustion.

A piston cavity 11b having a curved shape is formed on the top surface 11a of the piston 11. The piston cavity 11b is formed at a position where the tumble flow is easy to flow in, the curvature for turning the tumble flow in the direction of the intake pent roof surface 2e without difficulty, and a position where the injector 13 faces the fuel injection direction.

Further, an ignition plug 18 is provided for each cylinder, and a discharge electrode 18a of the ignition plug 18 is exposed between the intake ports 2a on the intake side pent roof surface 2e, and a tumble flow from the piston cavity 11b and It is located at a position facing the outflow direction of the fuel spray. That is, the fuel injection direction by the injector 13 is not directly directed to the discharge electrode 18a of the ignition plug 18 and is arranged.

An igniter 20 is connected to the ignition plug 18 via an ignition coil 19 provided for each cylinder.

The exhaust system of the engine 1 is shown in FIG.
As shown in FIG. 5, an exhaust pipe 22 is communicated with a collection portion of the exhaust manifold 21 which communicates with each exhaust port 2b of the cylinder head 2, and a catalytic converter 23 is interposed in the exhaust pipe 22 and communicated with a muffler 24. ing.

Next, the fuel system will be described. A fuel line 25 in FIG. 15 is provided with a high-pressure fuel pump 26 in the middle of the fuel line 25, and a high-pressure electromagnetic pressure regulator 27 downstream of the high-pressure fuel pump 26. ing.

An upstream side of the high-pressure fuel pump 26 of the fuel line 25 constitutes a low-pressure line 25a for sending fuel from a fuel tank 29 by a feed pump 28,
A high-pressure line 25b between the downstream side of the high-pressure fuel pump 26 and the high-pressure electromagnetic pressure regulator 27 constitutes a high-pressure line 25b that boosts the fuel from the low-pressure line 25a and supplies the fuel to the injector 13. The downstream side from the pressure regulator 27 constitutes a return line 25c for returning surplus fuel to the fuel tank 29.

The high-pressure electromagnetic pressure regulator 27 is a normally open type and has an electronic control unit 50 (see FIG.
2), the valve opening is reduced and the fuel pressure is increased as the duty ratio of the control signal output from the electronic control unit 50 is increased. That is, in the present embodiment, the fuel pressure is feedback-controlled by the electronic control unit 50, and the high-pressure line 2 is controlled by the high-pressure electromagnetic pressure regulator 27.
By adjusting the fuel relief amount of the high pressure line 5b, the fuel pressure in the high pressure line 25b is maintained at a predetermined pressure, and the amount of fuel injected into the combustion chamber 12 can be accurately measured based on the valve opening time of the injector 13.

The low pressure line 25a downstream of the feed pump 28 and the return line 25c are communicated via a fuel bypass passage 25d.
At 5d, a low-pressure diaphragm-type pressure regulator 30 that regulates the fuel pressure in the low-pressure line is interposed.

Then, the fuel in the fuel tank 29 is sent out by the feed pump 28 interposed in the low-pressure line 25a, passes through the fuel filter 31, and is regulated by the low-pressure diaphragm type pressure regulator 30. To the fuel pump 26 for use. The fuel from the low-pressure line 25a is pressurized by the high-pressure fuel pump 26, and a predetermined high-pressure fuel regulated by the high-pressure electromagnetic pressure regulator 27 is buffered by the high-pressure fuel filter 32 to reduce the pulsating pressure. The fuel is supplied to the injector 13 of each cylinder via a common rail 34 provided with an accumulator 33.

In the present embodiment, the high-pressure fuel pump 26 is an engine driven plunger pump.
A check valve is provided at each of the suction port and the discharge port (not shown). When the engine is stopped, the high-pressure electromagnetic pressure regulator 27 is fully opened by a normally-open type, and a low-pressure line is formed. 25a, high pressure line 25
The fuel pressure at the point b is released to the fuel tank 29, thereby preventing a problem such as fuel leakage from the injector 13 to the combustion chamber 12 due to holding of the high-pressure fuel when the engine is stopped.

Next, sensors for detecting the operating state of the engine will be described. The above throttle chamber 5
A throttle sensor 35 having a built-in throttle opening sensor 35a and an idle switch 35b that is turned on when the throttle valve 5a is fully closed is connected to the throttle valve 5a.

The cylinder block 1a of the engine 1
A knock sensor 36 is attached to the cooling water passage 3 communicating the left and right banks of the cylinder block 1a.
7, a cooling water temperature sensor 38 is provided, and a fuel pressure sensor 39 for detecting the fuel pressure in the high pressure line 25b is provided on the common rail 34.

The crankshaft 40 of the engine 1
A crank angle sensor 42 is provided on the outer periphery of a crank rotor 41 which is axially mounted on the cam shaft 43. Further, a cam rotor 44 connected to a cam shaft 43 which makes a half turn with respect to the crank shaft 40 has a cam angle for cylinder discrimination. A sensor 45 is provided opposite.

As shown in FIG. 20, the crank rotor 41 has projections 41a, 41b and 41c formed on the outer periphery thereof, and these projections 41a, 41b and 41c are connected to the cylinders (# 1 and # 2 cylinders). (# 3, # 4 cylinders) before compression top dead center (BTDC) θ1, θ2, θ3.
In this embodiment, θ1 = 97 ° CA, θ2 = 65 ° C
A, θ3 = 10 ° CA.

As shown in FIG. 21, on the outer periphery of the cam rotor 44, protrusions 44a, 44b,
A projection 44a is formed at a position θ4 after the compression top dead center (ATDC) of the # 3 and # 4 cylinders.
b is composed of three protrusions, and the first protrusion is A of the # 1 cylinder.
It is formed at the position of TDCθ5. Further, the protrusion 44c
Is formed by two projections, and the first projection is the AT of the # 2 cylinder.
It is formed at the position of DCθ6. In this embodiment,
θ4 = 20 ° CA, θ5 = 5 ° CA, θ6 = 20 ° CA.

As shown in the time charts of FIGS. 12 and 14, the crank rotor 41 and the cam rotor 44 are rotated by the rotation of the crankshaft 40 and the camshaft 43 in association with the operation of the engine. Is detected by the crank angle sensor 42, and θ1, θ2, θ3 (B
Each crank pulse of TDC 97 °, 65 °, 10 °) is output for each half engine revolution (180 ° CA).
On the other hand, each protrusion of the cam rotor 44 is detected by the cam angle sensor 45 between the θ3 crank pulse and the θ1 crank pulse, and the cam angle sensor 45 outputs a predetermined number of cam pulses.

As described later, in the electronic control unit 50,
The engine speed NE is calculated based on the input interval time Tθ of the crank pulse output from the crank angle sensor 42, and the order of the combustion strokes of each cylinder (for example, # 1 cylinder → # 3 cylinder → # 2 cylinder → Based on the pattern of the # 4 cylinder) and a value obtained by counting the cam pulse from the cam angle sensor 45 by a counter, the cylinders of the fuel injection target cylinder and the ignition target cylinder are determined.

The injector 13, the spark plug 18,
Calculation of control amounts for actuators such as the ISC valve 10 and output of control signals, that is, engine control such as fuel injection control, ignition timing control, idle speed control, etc.
The electronic control unit (ECU) 50 shown in FIG.

The ECU 50 comprises a CPU 51, a ROM 5
2, RAM53, backup RAM54, counter
A timer group 55 and an I / O interface 56 are mainly configured by a microcomputer connected to each other via a bus line, and are connected to the constant voltage circuit 57 for supplying a stabilized power to each unit, and to the I / O interface 56. And a peripheral circuit such as an A / D converter 59.

The counter / timer group 55 includes various counters such as a free-run counter, a counter for counting the input of a cam angle sensor signal (cam pulse), a fuel injection timer, an ignition timer, and a timer for generating a periodic interrupt. Various timers such as a periodic interrupt timer, a timer for measuring the input interval of a crank angle sensor signal (crank pulse), and a watchdog timer for monitoring a system abnormality are collectively referred to for convenience. Is used.

The constant voltage circuit 57 is connected to a battery 61 through a first relay contact of a power supply relay 60 having two circuit relay contacts, and a relay coil of the power supply relay 60 is connected to an ignition switch 62 by the battery 61.
Connected through. Further, the constant voltage circuit 57
Is connected directly to the battery 61. When the ignition switch 62 is turned on and the contact of the power supply relay 60 is closed, power is supplied to each part in the ECU 50, while the ignition switch 62 is turned on and off. Instead, the backup power is always supplied to the backup RAM 54. Further, the feed pump 28 is connected to the battery 61 via a relay contact of a feed pump relay 63. The second relay contact of the power supply relay 60 is connected to the battery 6.
1 is connected to a power supply line for supplying power to each actuator.

The input port of the I / O interface 56 includes an idle switch 35b and a knock sensor 3
6, a crank angle sensor 42, a cam angle sensor 45, a vehicle speed sensor 46 for detecting a vehicle speed, and a throttle opening sensor 35a, a coolant temperature sensor 38 via the A / D converter 59. , And the fuel pressure sensor 39 are connected, and the battery voltage VB is inputted and monitored.

On the other hand, the output ports of the I / O interface 56 are connected to the ISC valve 10, the injector 13, the high-pressure electromagnetic pressure regulator 27, and the relay coil of the feed pump relay 63 by the drive circuit 58.
And the igniter 20 is connected.

In accordance with the control program stored in the ROM 52, the CPU 51 processes the detection signals from the sensors / switches input via the I / O interface 56, the battery voltage, and the like, and stores them in the RAM 53. Based on various data, various learning value data stored in the backup RAM 54, fixed data stored in the ROM 52, and the like, a fuel injection amount, an ignition timing, a duty ratio of a drive signal for the ISC valve 10, and the like are calculated. It performs engine control such as injection control, ignition timing control, and idle speed control.

In such an engine control system, EC
In U50, an ignition fuel injection amount capable of forming a mixture that can be ignited by the ignition plug 18 at a minimum, and a main fuel injection amount for obtaining a predetermined output according to an engine operating state are set. The time or crank angle from when the fuel of the ignition fuel injection amount is injected to when the mixture of the injected fuel reaches the spark plug 18 after the injected fuel collides with the piston cavity 11b of the piston top surface 11a is determined. ,
Back calculation is performed based on the ignition timing, and the time or crank angle before the ignition is set as the ignition fuel injection timing. Further, a main fuel injection timing for injecting fuel based on the main fuel injection amount after the ignition fuel injection is set based on an operation state.

That is, the fuel is injected twice per one cylinder cycle, and as the first fuel injection, the amount of fuel that can form the minimum ignitable fuel-air mixture is determined based on the ignition timing. After colliding with the piston cavity 11b formed on the top surface 11a, the fuel is injected from the injector 13 at a time or a crank angle before ignition until reaching the discharge electrode 18a of the ignition plug 18.
Accordingly, the fuel injection direction of the injector 13 is arranged without directing the discharge electrode 18a of the spark plug 18 directly, and the fuel spray from the injector 13 collides with the piston cavity 11b of the piston top surface 11a, thereby in-cylinder air intake. When the air-fuel mixture in the flammable range that has rebounded reaches the discharge electrode 18a of the ignition plug 18 due to the synergistic action with the flow, the mixture can be ignited by the ignition plug 18 to prevent smoldering. While doing
Improves ignitability.

Thereafter, as the second fuel injection, the fuel of the main fuel injection amount for obtaining a predetermined output according to the engine operating state is injected at a timing corresponding to the operating state. As a result, the main fuel injection can be performed immediately before ignition, and the combustion temperature can be reduced to reduce the generation of NOx. In addition, since the flame propagates to the air-fuel mixture formed by the second injected fuel due to the ignition of the air-fuel mixture by the first injected fuel, even in the lean air-fuel mixture as a whole in the combustion chamber 12, in the portion where combustion is performed, Therefore, it is possible to form an air-fuel mixture in the flammable range, so that the lean limit can be greatly improved, and the fuel efficiency can be improved.

As a result, smoldering can be prevented, and it is possible to achieve both improvement in ignitability and improvement in exhaust emission and fuel efficiency by improving the air-fuel ratio lean limit.

More specifically, fuel injection is performed twice per cylinder during stratified charge combustion, and once per normal cylinder per cycle during uniform mixed combustion. That is, stratified charge combustion is selected at low rotation and low load, and uniform mixed combustion is selected at high rotation and high load based on the engine operating state. When the stratified combustion is selected, the ignition fuel injection pulse width TiFST for obtaining the ignition fuel injection amount GFFST capable of forming the mixture ignitable by the spark plug 18 at a minimum, and the stratified combustion in accordance with the engine operating state. And a main fuel injection pulse width TiMAIN for obtaining a main fuel injection amount GFMAIN for obtaining a predetermined output. The time or crank angle between the time when the fuel of the ignition fuel injection amount is injected and the time when the mixture of the injected fuel reaches the spark plug 18 after the injected fuel collides with the piston top surface 11a is defined as the ignition timing. The time before the ignition or the crank angle is set as the ignition fuel injection timing. Further, a main fuel injection timing for injecting fuel according to the main fuel injection amount after the ignition fuel injection is set based on the engine operating state, adapted to stratified combustion. The fuel injection is performed twice per one cylinder cycle, and as the first fuel injection, an ignition fuel injection amount is set so as to minimize the mixture ignitable by the ignition plug, and the fuel injection amount of this ignition fuel injection amount is set. The time or crank angle from when the injected fuel collides with the top surface of the piston until the air-fuel mixture of the injected fuel reaches the spark plug is calculated back with reference to the ignition timing. While the angle is set as the ignition fuel injection timing, when uniform mixed combustion is selected, the fuel injection pulse width Ti for obtaining the fuel injection amount GF suitable for uniform mixed combustion is set based on the engine operating condition. Then, the fuel injection timing for injecting the fuel with the above fuel injection amount adapted to the uniform mixed combustion is set according to the engine operating state.

That is, the ECU 50 sets the ignition fuel injection amount setting means, the main fuel injection amount setting means according to the present invention,
The functions of the ignition fuel injection timing setting unit and the main fuel injection timing setting unit are realized, and further, the functions of the combustion method selection unit, the fuel injection amount setting unit, and the fuel injection timing setting unit are also realized.

Hereinafter, the control process according to the present invention executed by the ECU 50 will be described with reference to the flowcharts shown in FIGS.

First, the ignition switch 62 is turned on.
Then, when the power is supplied to the ECU 50, the system is initialized, and each flag and each counter are initialized except for data such as various learning values stored in the backup RAM 54. Then, when a starter switch (not shown) is turned on and the engine 1 is started, a cylinder discriminating / engine speed calculating routine shown in FIG. 2 is executed every time a crank pulse is input from the crank angle sensor 42.

In this cylinder discriminating / engine speed calculating routine, when the crank rotor 41 rotates in response to engine operation and a crank pulse is input from the crank angle sensor 42, first, in step S1, the presently input crank is input. Whether the pulse corresponds to the crank angle of θ1, θ2, or θ3 is determined based on the input pattern of the cam pulse from the cam angle sensor 45. In step S2, the cylinder to be ignited is determined from the input pattern of the crank pulse and the cam pulse. Then, a cylinder such as a fuel injection target cylinder is determined.

That is, as shown in the time charts of FIGS. 12 and 14, for example, if there is a cam pulse input between the previous crank pulse input and the present crank pulse input, the present crank pulse is The crank pulse can be identified as a θ1 crank pulse, and the crank pulse input next time can be identified as a θ2 crank pulse.

When there is no cam pulse input between the previous and present crank pulse inputs and there is a cam pulse input between the last and previous crank pulse inputs, the present crank pulse can be identified as the θ2 crank pulse, and The input crank pulse can be identified as a θ3 crank pulse. Further, when there is no cam pulse input between the previous and current times and between the last and last crank pulse inputs, the currently input crank pulse can be identified as the θ3 crank pulse, and the next input crank pulse is It can be identified as a θ1 crank pulse.

Further, three cam pulses are input between the previous and current crank pulse inputs (θ corresponding to the projection 44b).
When 5 cam pulses are applied, the next compression top dead center is the # 3 cylinder, the cylinder to be ignited can be determined to be the # 3 cylinder, and the next fuel injection target cylinder can be determined to be the next cylinder # 2. When two cam pulses are input between the previous and current crank pulse inputs (θ6 cam pulse corresponding to the projection 44c), the next top dead center of the compression is # 4 cylinder,
The ignition target cylinder is # 4 cylinder, the next fuel injection target cylinder is #
It can be determined as one cylinder.

If one cam pulse is input between the previous and current crank pulse inputs (.theta.4 cam pulse corresponding to the projection 44a) and the previous compression top dead center determination is for the # 4 cylinder, the next compression The dead center is cylinder # 1, the cylinder to be ignited can be determined to be cylinder # 1, and the cylinder to be fueled next time can be determined to be cylinder # 3. Similarly, one cam pulse is input between the previous and current crank pulse inputs, and the previous compression top dead center determination is #
When the number of cylinders is three, the next compression top dead center is cylinder # 2, the cylinder to be ignited is cylinder # 2, and the cylinder to be fueled next time is cylinder # 4.

Thereafter, the process proceeds from step S2 to step S3, in which the time from the previous crank pulse input to the present crank pulse input measured by the crank pulse input time counting timer, that is, the crank pulse input interval time (θ1 crank pulse And the input interval time Tθ12 of the θ2 crank pulse, the input interval time Tθ23 of the θ2 crank pulse and the θ3 crank pulse, or the input interval time Tθ31 of the θ3 crank pulse and the θ1 crank pulse), and the crank pulse input interval time Tθ is detected.

Then, the program proceeds to a step S4, wherein an inter-crank pulse angle corresponding to the presently identified crank pulse is read, and a current engine speed NE is calculated based on the crank pulse angle and the crank pulse input interval time. Is stored at a predetermined address in the RAM 53 and the processing exits from the routine. The angle between the crank pulses is known and is stored in advance in the ROM 52 as fixed data. In this embodiment, the angle θ12 between the θ1 crank pulse and the θ2 crank pulse is 32 ° CA, and Angle θ between crank pulse and θ3 crank pulse
23 is 55 ° CA, and the angle θ 31 between the θ3 crank pulse and the θ1 crank pulse is 93 ° CA.

In the ignition control routine shown in FIG. 3 and the fuel injection control routine shown in FIGS. 4 and 5 which are executed at predetermined intervals, the engine speed NE is read, and the ignition timing and the fuel injection pulse width are read. Is used when setting the fuel injection timing.

First, prior to the description of the fuel injection control routine, the ignition control routine of FIG. 3 will be described.

This ignition control routine is executed at predetermined intervals (for example, at 180 ° CA).
Then, an area determination value αTGT for determining whether to select between stratified combustion and uniform mixed combustion is set by referring to the area determination value table with interpolation calculation based on the current engine speed NE. .

The region determination value αTGT is a reference value when switching the combustion mode according to the engine load. In this embodiment, the throttle opening α is adopted as an example of the engine load, and the throttle opening α The area determination value αTGT
By comparing with the above, it is determined whether to select stratified combustion or uniform mixed combustion. The switching of the combustion method is performed by changing the fuel injection timing and the ignition timing.

Here, the stratified combustion is a combustion system in which the fuel injection is terminated immediately before ignition and the rear end of the fuel spray is ignited by the spark plug 18, and only the air around the fuel is used. Since it is possible to obtain stable combustion with a very small amount of fuel, it is suitable for operation in a low rotation and low load region. On the other hand, homogeneous mixed combustion is a combustion method in which fuel is injected at a relatively early time, and the injected fuel and air are mixed uniformly in a cylinder and then ignited. Suitable for driving.

Accordingly, the above-mentioned region determination value table determines in advance the appropriate throttle opening for switching between stratified combustion and uniform mixed combustion by simulation or experiment or the like for each engine speed region, and calculates the appropriate throttle opening in the region determination value αTGT. Is set as a table using the engine speed NE as a parameter, and is stored in a series of addresses in the ROM 52. As shown in FIG. The value αTGT is stored.

Then, in step S12, the current throttle opening α by the throttle opening sensor 35a is compared with the above-mentioned region determination value αTGT set by referring to the table based on the engine speed NE, and a low rotation speed α ≦ αTGT is obtained. At the time of load (shaded area in FIG. 11), stratified combustion is selected to improve fuel efficiency and exhaust emission, and α> αTGT
At high engine speed and high load, uniform mixed combustion is selected to secure engine output.

When stratified charge combustion is selected, step S1
Proceed to 3 to set a stratified combustion selection flag FSTR indicating that stratified combustion has been selected (FSTR ← 1), and then proceed to the next step.
At S14, the basic ignition timing table for stratified combustion stored in the ROM 52 is referenced with interpolation calculation based on the engine speed NE and the throttle opening α, and the basic ignition timing as a basic ignition timing adapted to stratified combustion is determined. Set the advance value ADVBASE.

The stratified combustion basic advance value table includes a throttle opening α indicating an engine load and an engine speed N.
The optimum ignition timing suitable for stratified combustion is obtained in advance by simulation or experiment, etc., for each region according to E, and the optimal ignition timing suitable for this stratified combustion is set as a basic advance value ADVBASE that determines at what CA the BTDC should be ignited. , The throttle opening α and the engine speed NE are set as a table using parameters as parameters and stored in a series of addresses in the ROM 52.

Thereafter, the process proceeds to step S15, in which the ignition timing learning correction value ADVKR for learning the retard or advance amount for each operation region in accordance with the presence or absence of knock detected by the knock sensor 36 is set to the throttle opening α. Based on the engine speed, the ignition timing learning correction value table stored in the backup RAM 54 is set with reference to interpolation calculation.

Then, the process proceeds to step S16, in which the ignition timing learning correction value ADVKR is added to the basic advance value ADVBASE to perform learning correction, and a control advance angle ADV that determines the ignition timing is set (ADV ← ADVBASE + ADVKR).

Next, at step S17, the control advance AD
Based on V, an energization cutoff timing for the ignition coil 19 based on the θ1 crank pulse input, that is, an ignition timing TADV is set.

In the present embodiment, the ignition timing is controlled by a so-called time control method. As shown in FIGS. 12 and 14, the energization start timing (dwell set) and the energization cutoff timing (ignition (Timing; dwell cut) is set by the time after the input of the θ1 crank pulse.

That is, since the control advance ADV is angle data (BTDC ° CA), it is necessary to convert the control advance ADV into the time from the input of the θ1 crank pulse to the ignition.
Assuming that the latest crank pulse input interval time is Tθ and the angle between the crank pulses corresponding to the latest crank pulse input interval time Tθ is θ, in the present embodiment, the ignition timing TADV is calculated based on the θ1 crank pulse input.
The time is set by the following equation from the time per 1 ° CA rotation (Tθ / θ).

TADV ← (Tθ / θ) × (θ1−ADV) θ1; In the present embodiment, θ1 = 97 ° CA Then, in step S18, the ignition coil is referred by referring to the table based on the battery voltage VB with interpolation calculation. The energization time (dwell) DWL for 19 is set. This energizing time D
WL determines the optimum energizing time of the coil primary current depending on the battery voltage, and an example of this table is shown in step S18. That is, when the battery voltage VB decreases, the energizing time DWL is lengthened to secure ignition energy, and when the battery voltage VB increases, the energizing time DWL
To prevent energy loss and heat generation of the ignition coil 19.

Next, in step S19, the energization time DWL is subtracted from the ignition timing TADV to set the energization start timing TDWL based on the θ1 crank pulse input (TDWL ← TADV-DWL). The ignition timing TADV adapted to the stratified combustion is set in the ignition timing timer of, and in the following step S21, the energization start timing TDWL is set in the energization start timing timer of the corresponding cylinder, and the routine exits.

On the other hand, in step S12, α> αT
When the current engine operating state of the GT is a high-speed high-load and uniform mixed combustion is selected, the process proceeds from step S12 to step S22,
The stratified combustion selection flag FST is selected by selecting the uniform mixed combustion.
R is cleared (FSTR ← 0), and in the following step S23, the ROM is determined based on the engine speed NE and the throttle opening α.
The basic advance value ADVBASE as the basic ignition timing suitable for uniform mixed combustion is set by referring to the basic advance value table for uniform mixed combustion stored in 52 with interpolation calculation.

The basic advance value table for uniform mixed combustion is obtained by simulating or experimenting in advance the optimum ignition timing suitable for uniform mixed combustion for each region based on the throttle opening α representing the engine load and the engine speed NE. The optimum ignition timing suitable for this homogeneous mixed combustion is set as a basic advance value ADVBASE as a table using the throttle opening α and the engine speed NE as parameters, and stored in a series of addresses in the ROM 52. is there.

Then, the process proceeds to step S15, where the ignition timing learning correction value ADVKR is set by referring to the ignition timing learning value correction table with interpolation calculation based on the throttle opening α and the engine speed NE. S16, S17
Then, the ignition timing learning correction value A is added to the basic advance value ADVBASE adapted to uniform mixed combustion set in step S23.
The control advance angle ADV is set by adding DVKR, and the control advance angle ADV based on the angle data is converted into the ignition timing TADV based on the θ1 crank pulse input. Further, in steps S18 and S19, the energizing time DWL is set by referring to the table based on the battery voltage VB, and the energizing start timing TDWL is set by subtracting the energizing time DWL from the ignition timing TADV, and the steps S20 and S2 are performed.
In step 1, the ignition timing TADV adapted to the uniform mixed combustion is set in the ignition timing timer of the corresponding cylinder, the energization start timing TDWL is set in the energization start timing timer of the cylinder, and the routine exits.

As a result, the above timers are started by a .theta.1 crank pulse interruption routine of FIG. 6 which is started in synchronization with the .theta.1 crank pulse input, and ignition suitable for each combustion mode is performed.

In the fuel injection control routine shown in FIGS. 4 and 5, the ignition timing T for determining the ignition timing is set.
ADV is referred to, and a fuel injection timing that determines the fuel injection timing of the corresponding cylinder is set based on the ignition timing TADV.

Next, the fuel injection control routine shown in FIGS. 4 and 5 will be described.

This fuel injection control routine is executed at predetermined intervals (for example, 180 ° CA).
In S31, the fuel pressure coefficient KS is set by referring to the table with interpolation calculation based on the fuel pressure PFUEL (fuel pressure in the high pressure line) to the injector 13 detected by the fuel pressure sensor 39.

The fuel pressure coefficient KS is equal to the fuel pressure PFUEL
In addition to correcting the injection characteristics of the injector 13 that change due to the above, the fuel injection amount (unit: g) described later is converted into time. An example of the table is shown during step S31. If the valve opening time of the injector 13 is the same, the fuel injection amount increases as the fuel pressure PFUEL increases. Therefore, even if the same fuel injection pulse width is given to the injector 13, the fuel injection amount is reduced due to the difference in fuel pressure. Will change. For this reason, the fuel pressure P
As the FUEL increases, a lower value of the fuel pressure coefficient KS is stored.

Next, at step S32, the fuel pressure PFUE
The invalid injection pulse width TS is set by referring to the table with interpolation calculation based on L. This invalid injection pulse width TS
Is for compensating for the operation delay of the injector 13 which varies with the fuel pressure PFUEL.
An example of the table is shown in S32.

As the fuel pressure PFUEL is higher, the injector 13 is opened against this fuel pressure. Therefore, the table stores the invalid injection pulse width Ts of a longer time value.

Thereafter, the flow proceeds to step S33, where the stratified combustion selection flag FSTR is referred to.

When the stratified charge combustion is selected with FSTR = 1, the routine proceeds to step S34, where the fuel mixture is ignited by the ignition plug 18 by the processing of steps S34 to S45 in order to perform the fuel injection twice per one cylinder cycle. The main fuel injection pulse width TiFST for obtaining the ignition fuel injection amount GFFST that can form the minimum fuel injection amount, and the main fuel injection amount GFMAIN for adapting to the stratified combustion and obtaining a predetermined output according to the engine operating state. The fuel injection pulse width TiMAIN is set. The time or crank angle between the time when the fuel of the ignition fuel injection amount is injected and the time when the mixture of the injected fuel reaches the spark plug 18 after the injected fuel collides with the piston top surface 11a is defined as the ignition timing. Back calculation is performed with reference to the determined ignition timing TADV, and the time or crank angle before the ignition is set as the ignition fuel injection timing, that is, the ignition fuel injection start timing IJSTFST. In addition, the main fuel injection start timing IJSTMAIN, which is a main fuel injection timing adapted to the stratified combustion and injecting the fuel by the main fuel injection amount after the ignition fuel injection, is set to the engine speed NE and the throttle opening α. Set based on:

In step S 34, the ignition fuel injection amount table is referred to with interpolation calculation based on the engine speed NE and the throttle opening α, and ignition is performed so that the air-fuel mixture ignitable by the spark plug 18 can be formed at a minimum. Fuel injection amount GFFS
Set T.

The fuel injection amount table for ignition stores fuel spray from the injector 13 on the piston top surface 11a by simulation or experiment in advance for each region based on the throttle opening α representing the engine load and the engine speed NE. The air-fuel ratio air-fuel mixture which ignites by the spark plug 18 to the air-fuel mixture which has rebounded by the synergistic action with the in-cylinder intake air flow (tumble flow) colliding with the piston cavity 11b is formed at a minimum. Is determined, and this fuel injection amount is used as the ignition fuel injection amount GF
The FST is set as a table using the throttle opening α and the engine speed NE as parameters and stored in the ROM 52.

In step S35, the main fuel injection amount table is referred to with interpolation calculation based on the engine speed NE and the throttle opening α to obtain a main output for adapting to stratified combustion and obtaining a predetermined output. Set the fuel injection amount GFMAIN.

The main fuel injection amount table is stratified in consideration of the ignition fuel injection amount by simulation or experiment in advance for each region based on the throttle opening α representing the engine load and the engine speed NE. An optimum fuel injection amount for adapting to combustion and obtaining a predetermined output is obtained, and this optimum fuel injection amount is set as a main fuel injection amount GFMAIN, and is set as a table using a throttle opening α and an engine speed NE as parameters, It is stored in the ROM 52.

In the following step S36, the engine speed NE
The ignition fuel injection end timing table is searched based on the throttle opening α and the ignition fuel injection end timing IJENDFST (unit: msec) is set by interpolation calculation.

The ignition fuel injection end timing table shows that the fuel spray from the injector 13 after the fuel injection has been performed in advance by simulation or experiment for each region based on the throttle opening α representing the engine load and the engine speed NE. The air-fuel mixture colliding with the piston cavity 11b on the piston top surface 11a and synergizing with the in-cylinder intake air flow (tumble flow) causes the rebounded air-fuel mixture to generate a spark plug 1
8 is obtained, and this time value is set as the ignition fuel injection end timing IJENDFST as a table using the throttle opening α and the engine speed NE as parameters, and is stored in the ROM 52. .

That is, the ignition fuel injection end timing IJENDFST is the ignition timing TADV (ignition timing).
It is determined how many milliseconds before the ignition fuel injection is to end.

In the present embodiment, the ignition fuel injection end timing IJENDFST is set by a time value. However, the ignition fuel injection end timing IJENDFST is set by an angle value (unit: ° CA) based on the ignition timing. It may be determined beforehand whether to end the fuel injection for ignition.

Next, at step S37, the engine speed NE
The main fuel injection end timing table is searched based on the throttle opening α and the main fuel injection end timing IJENDMAIN (unit: msec) is set by interpolation calculation.

In the stratified combustion in the present embodiment, as described above, the fuel spray from the fuel injection for ignition from the injector 13 is applied to the piston cavity 1 on the piston top surface 11a.
1b, and when the air-fuel ratio mixture in the combustible range that has rebounded reaches the discharge electrode 18a of the ignition plug 18 due to the synergistic action with the in-cylinder intake flow, the ignition of the ignition plug 18 The fuel mixture is ignited by the first injection of fuel and the flame is propagated to the rear end of the fuel spray formed by the second injection of fuel by the main fuel injection. Need to manage time intervals.

Therefore, the optimal end timing of the main fuel injection before ignition adapted to the stratified combustion is determined for each region by the throttle opening α representing the engine load and the engine speed NE.
The optimum fuel injection end timing of the main fuel injection is determined in advance by simulation or experiment, and is set as a main fuel injection end timing IJENDMAIN as a main fuel injection end timing table using the throttle opening α and the engine speed NE as parameters. It is stored in the ROM 52.

Then, the process proceeds to a step S38, in which the ignition fuel injection amount GFFST set in the step S34 is multiplied by the fuel pressure coefficient KS to convert the time, and the value is added to the invalid injection pulse width TS. , Ignition fuel injection amount GFFST
Is calculated (TiFST ← KS × GFFST + TS).

At the following step S39, the fuel pressure coefficient K is added to the main fuel injection amount GFMAIN set at step S35.
In addition to multiplying the time by S, the time is converted, and the correction is performed by multiplying by various correction coefficients COEF set in accordance with the operation state, and the value is added to the invalid injection pulse width TS to adapt to the stratified combustion. Main fuel injection pulse width Ti for obtaining a fuel injection amount for obtaining a predetermined output according to the engine operating state
Calculate MAIN (TiMAIN ← KS × GFMAIN × COEF
+ TS).

In step S40, the fuel injection end timing IJENDFST set in step S36 and the ignition fuel injection pulse TiFST are calculated backward from the ignition timing TADV of the corresponding cylinder calculated in the ignition control routine. , The previous θ1 for the cylinder
Ignition fuel injection start timing IJSTFST that determines ignition fuel injection start timing based on crank pulse input
(Unit: msec) is calculated.

In the present embodiment, the fuel injection start timing is controlled by the time control method. As shown in the time chart of the relationship between the ignition signal and the injector drive signal during stratified combustion in FIG. The injection start timing is set based on the time after the input of the immediately preceding θ1 crank pulse for the corresponding cylinder.

That is, the ignition fuel injection end timing IJENDFST represents a time interval between the end of the ignition fuel injection and the ignition, and is a time value before the ignition. Therefore, it is necessary to convert this to the time from the input of the θ1 crank pulse to the start of the injection of the ignition fuel, and as shown in FIG. The fuel injection start timing must be obtained at the time of the pulse input.

For this reason, in the present embodiment, the ignition fuel injection start timing IJSTFST for the relevant cylinder is calculated backward from the ignition timing TADV of the relevant cylinder with reference to the immediately preceding θ1 crank pulse input.

The latest crank pulse input interval time is T
θ, the angle between the crank pulses corresponding to the latest crank pulse input interval time Tθ is θ, θ1 at which the fuel injection start timing starts counting, and θ1 of the corresponding cylinder as the reference for setting the ignition timing TADV from the crank pulse input Time Tθ11 until crank pulse input is 1 ° C
From the time per A rotation (Tθ / θ), it can be calculated by the following equation.

Tθ11 = (Tθ / θ) × θ11 θ11; The angle between θ1 crank pulses. In the present embodiment, θ11 = 180 ° CA. By subtracting the ignition fuel injection end timing IJENDFST from the value obtained by adding TADV, and further subtracting the ignition fuel injection pulse width TiFST, the ignition fuel injection start timing IJSTFST for the corresponding cylinder is calculated.

IJSTFST ← (Tθ / θ) × θ11 + TADV−
(TiFST + IJENDFST) Here, the ignition fuel injection end timing IJEN
If DFST is given by crank angle data before ignition based on ignition timing,
The ignition fuel injection start timing IJSTFST is calculated.

IJSTFST ← (Tθ / θ) × θ11 + TADV−
(TiFST + IJENDFST × (Tθ / θ)) Next, in step S41, the main fuel injection set in step S37 from the time Tθ11 (= (Tθ / θ) × θ11) between the θ1 crank pulses and the ignition timing TADV. The main fuel injection start timing IJSTMAIN (unit: msec) which determines the start timing of the main fuel injection based on the input of the immediately preceding θ1 crank pulse for the corresponding cylinder by calculating backward with the end timing IJENDMAIN and the main fuel injection pulse TiMAIN. Is calculated by the following equation.

IJSTMAIN ← (Tθ / θ) × θ11 + TADV
− (TiMAIN + IJENDMAIN) Although the controllability is poor, the engine speed NE
And the throttle opening α representing the engine load,
The ignition fuel injection start timing and the main fuel injection start timing are given with reference to the ignition timing TADV, and the ignition fuel injection start timing IJSTFST and the main fuel injection start timing are calculated back from the ignition timing TADV using these fuel injection start timings. The injection start timing IJSTMAIN may be set.

Then, the process proceeds to a step S42, wherein the ignition fuel injection pulse width TiFST calculated in the step S38 is set in the first fuel injection timer of the corresponding cylinder. In a succeeding step S43, the main fuel calculated in the step S39 is calculated. The injection pulse width TiMAIN is set in the second fuel injection timer of the cylinder.

In step S44, the ignition fuel injection start timing IJSTFS calculated in step S40 is calculated.
T is set in the first injection start timing timer of the relevant cylinder, and in the following step S45, the main fuel injection start timing IJSTMAIN calculated in the above step S41 is set in the second injection start timing timer of the relevant cylinder,
Exit the routine.

On the other hand, in step S33, FSTR =
If 0, that is, if the uniform mixed combustion is selected, the process proceeds to step S46 in FIG. 5, and a process for injecting fuel once per cylinder per cycle is performed as usual. That is, when uniform mixing combustion is selected, the fuel injection pulse width Ti for obtaining the fuel injection amount GF suitable for uniform mixing combustion is set based on the engine operating state. Then, a fuel injection start timing IJST, which is a fuel injection timing for injecting fuel based on the above fuel injection amount, is set based on the engine speed NE and the throttle opening α in order to adapt to the uniform mixed combustion.

In step S46, referring to the fuel injection amount table with interpolation calculation based on the engine speed NE and the throttle opening α, the fuel injection suitable for uniform mixed combustion and obtaining a predetermined engine output. Set the volume GF.

The fuel injection amount table is prepared by simulation or experiment in advance for each area based on the throttle opening α representing the engine load and the engine speed NE.
An optimum fuel injection amount for obtaining a predetermined engine output adapted to uniform mixed combustion is determined, and this optimum fuel injection amount is set as a fuel injection amount GF, and set as a table using a throttle opening α and an engine speed NE as parameters. And RO
It is stored in M52.

In the following step S47, the engine speed NE
A fuel injection start angle table is searched based on the throttle opening α and the fuel injection start angle IJsa (unit: ° CA) is set by interpolation calculation with reference to the compression top dead center of the corresponding cylinder.

That is, in the case of uniform mixed combustion, it is desirable to end the fuel injection end timing as early as possible so that the fuel is sufficiently mixed with fresh air. The injection pulse width becomes longer and, contrary to this, the time required for one cycle becomes shorter. Therefore, if the fuel injection start timing is not properly managed, the fuel injection may be started from the exhaust stroke, and the fuel may flow through.

For this reason, after the end of the exhaust stroke, it is necessary to start the fuel injection, and this fuel injection start timing is managed by the crank angle based on the compression top dead center of the corresponding cylinder (see FIG. 14).

Accordingly, the optimum start angle of the fuel injection before the compression top dead center of the cylinder corresponding to the uniform mixed combustion is determined by the throttle opening α representing the engine load and the engine speed NE.
For each of the following regions, a fuel injection start angle table is obtained in advance by simulation or experiment, and the optimum fuel injection start angle of this fuel injection is set as the fuel injection start angle IJsa, and the throttle opening α and the engine speed NE are used as parameters. Are set and stored in the ROM 52.

Then, the process proceeds to a step S48, in which the fuel injection amount GF set in the step S46 is multiplied by the fuel pressure coefficient KS to convert the time, and is multiplied by various correction coefficients COEF set according to the operation state. The fuel injection pulse width is obtained by adding the invalid injection pulse width Ts to the value to obtain a fuel injection amount adapted to uniform mixed combustion and to obtain a predetermined engine output according to the engine operating state at that time. Calculate Ti (Ti ← KS × GF × COEF
+ TS).

Then, in step S49, the above-described step S47
Based on the fuel injection start angle IJsa set in
The fuel injection start timing IJST (unit: msec) is calculated as a time value that determines the fuel injection start timing after the input of the reference crank pulse.

In the present embodiment, as described above,
The fuel injection start timing is controlled by the time control method. As shown in the time chart of the relationship between the ignition signal and the injector drive signal at the time of uniform mixed combustion in FIG. 14, the fuel injection start timing IJST for the corresponding cylinder is performed.
Is set according to the time after the input of the immediately preceding θ1 crank pulse for the corresponding cylinder.

That is, the fuel injection start angle IJsa
Is the crank angle data based on the compression top dead center of the corresponding cylinder, which is converted into time, and this value is set to one for the corresponding cylinder used as a reference when setting the fuel injection start timing IJST. A desired fuel injection start timing IJST can be calculated by subtracting from the time from the previous θ1 crank pulse to the compression top dead center of the corresponding cylinder.

The latest crank pulse input interval time is T
θ, the angle between the crank pulses corresponding to the latest crank pulse input interval time Tθ is θ, and the corresponding cylinder as the reference for setting the fuel injection start timing IJST from the immediately preceding θ1 crank pulse input is The time TθS until reaching the compression top dead center can be calculated from the time per 1 ° CA rotation (Tθ / θ) by the following equation.

T.theta.S = (T.theta ./. Theta.). Times..theta.S .theta.S; The angle from the immediately preceding .theta.1 crank pulse to the compression top dead center of the relevant cylinder. In this embodiment, .theta.S = 180 + 97 = 277.degree. CA.

Therefore, the time value TθS (= (Tθ /
The fuel injection start timing IJST is calculated by subtracting the time-converted value of the fuel injection start angle IJsa from θ) × θS).

IJST ← (Tθ / θ) × (θS-IJsa) Then, the process proceeds to step S50, in which the fuel injection pulse width Ti calculated in step S48 is set in the first fuel injection timer of the corresponding cylinder, and the following steps are performed. In S51, the above step S
The fuel injection start timing IJST calculated in 49 is set in the first injection start timing timer of the corresponding cylinder, and the routine exits.

Here, in the case of the uniform mixed combustion, the fuel injection is performed once per one cylinder per cycle.
The second fuel injection timer and the second injection start timing timer are not used.

As a result, the timers except the fuel injection timer set in the ignition control routine and the fuel injection control routine are started by the θ1 crank pulse interruption routine of FIG. 6 which is started in synchronization with the input of the θ1 crank pulse. Then, ignition and fuel injection suitable for each combustion mode are performed.

Next, the θ1 crank pulse interruption routine shown in FIG. 6 will be described.

This θ1 crank pulse interruption routine is executed every time a θ1 crank pulse is input in association with the operation of the engine. In step S61, an ignition timing timer for the cylinder to be ignited is started. Starts the energization start timing timer.

Next, in step S63, the stratified combustion selection flag FSTR is referred to.

Then, when stratified charge combustion is selected with FSTR = 1, the routine proceeds to step S64, where the first injection start timing timer of the fuel injection target cylinder is started, and step S65 is started.
Then, the second injection start timing timer of the fuel injection target cylinder is started, and the routine exits.

That is, for example, as shown in the time chart of FIG. 12, if this routine is started by the BTDC θ1 pulse input of the # 1 cylinder, the ignition target cylinder is the # 1 cylinder, and the fuel injection target cylinder is # 3 cylinder.

Here, for the sake of simplicity, the same cylinder will be described as shown in FIGS. 12 and 13. When stratified charge combustion is selected, the immediately preceding θ1 crank before the compression top dead center of the cylinder is selected. By the pulse input, the first injection start timing timer and the second injection start timing timer of the corresponding cylinder are started.

When stratified charge combustion is selected, the ignition fuel injection start timing IJSTFST, which is set back from the ignition timing TADV by the fuel injection control routine, is set in the first injection start timing timer. Ignition fuel injection start timing IJSTFST
Is based on the ignition timing TADV, before the ignition fuel is injected before the ignition, and the fuel spray from the injector 13 is applied to the piston cavity 11 on the piston top surface 11a.
b, and the synergistic action with the in-cylinder intake air flow (tumble flow) gives time until the rebounded mixture reaches the spark plug 18.

The second injection start timing timer has an ignition timing T according to a fuel injection control routine.
The main fuel injection start timing IJSTMAIN, which is set by calculating backward from ADV, is set. This main fuel injection start timing IJSTMAIN is adapted to stratified combustion.
And the mixture is ignited by the first ignition injection fuel, so that an appropriate time is provided for propagating the flame to the rear end of the fuel spray formed by the second injection fuel by the main fuel injection. .

Then, the ignition fuel injection start timing IJ is determined by the timing of the first injection start timing timer.
When STFST is reached, the IJST1 interrupt routine shown in FIG. 7 is started, and in step S71, the first fuel injection timer of the corresponding cylinder is started, and the ignition fuel injection set in the first fuel injection timer is started. The injector drive signal based on the pulse width TiFST is applied to the injector 1 of the corresponding cylinder.
3 (see FIG. 12) and exits the routine.

As a result, as shown in FIG. 16, the fuel can be ignited from the injector 13 of the corresponding cylinder by the predeterminedly measured ignition plug 18 corresponding to the ignition fuel injection pulse width TiFST, as shown in FIG. Fuel is injected to obtain the ignition fuel injection amount GFFST that can form the air-fuel mixture at a minimum.

After that, the main fuel injection start timing IJSTMA is measured by the timing of the second injection start timing timer.
Upon reaching IN, an IJST2 interrupt routine shown in FIG. 8 is started, and in step S81, the second fuel injection timer of the corresponding cylinder is started, and the main fuel injection pulse set in the second fuel injection timer is set. An injector drive signal with the width TiMAIN is output to the injector 13 of the corresponding cylinder, and the routine exits.

As a result, as shown in FIG. 18, as the second fuel injection, the injector 13 of the corresponding cylinder adapts to the predetermined metered stratified charge combustion corresponding to the main fuel injection pulse width TiFST. Fuel is injected to obtain a fuel injection amount for ensuring a predetermined output according to the engine operating state.

Then, the BTDC θ1 crank pulse input of the relevant cylinder starts the ignition timing timer and the energization start timing timer of the relevant cylinder.

When the energization start timing timer reaches the energization start timing TDWL, the TD shown in FIG.
The WL interrupt routine is started, and in step S91, the igniter 20 is controlled from the ECU 50 by the dwell set of the corresponding cylinder.
An energization signal for the corresponding cylinder is output (see FIG. 12), and energization (dwell) of the ignition coil 19 of the corresponding cylinder is started.

Thereafter, when the timing of the ignition timing timer reaches the ignition timing TADV suitable for stratified combustion set in the ignition timing timer, a TADV interrupt routine shown in FIG. 10 is started, and in step S101, the corresponding cylinder is started. Cut the dwell for the ignition coil 19 and exit the routine.

As a result, a high secondary voltage is induced in the ignition coil 19 of the relevant cylinder, and the discharge electrode 18a of the spark plug 18 of the relevant cylinder sparks.

As described above, as the first fuel injection from the injector 13, the ignition fuel injection start timing IJSTFST is set back-calculated based on the ignition timing TADV, so that as shown in FIG. From the first ignition fuel injection collides with the piston cavity 11b on the piston top surface 11a, and synergizes with the in-cylinder intake flow (tumble flow).
When the rebounded mixture reaches the discharge electrode 18a of the ignition plug 18 without fail, the ignition plug 18 of the cylinder is ignited and the mixture is ignited and burned.

Thus, the fuel injection direction by the injector 13 is arranged without directing to the discharge electrode 18a of the ignition plug 18, and the first injection from the injector 13 is performed.
The fuel spray from the second ignition fuel injection is applied to the piston top surface 1
When the air-fuel mixture in the flammable range that has rebounded reaches the discharge electrode 18a of the ignition plug 18 due to a synergistic effect with the in-cylinder intake flow (tumble flow) when the air-fuel mixture collides with the piston cavity 11b of FIG. Spark plug 18 for mixture
Ignited, the fuel droplets are prevented from adhering to the discharge electrode 18a of the ignition plug 18, carbon is prevented from being deposited on the discharge electrode 18a, and smoldering is prevented while flammable The air-fuel mixture in the range can be surely present at the position of the discharge electrode 18a of the ignition plug 18, so that stable ignition can be obtained and ignitability can be improved.

The ignition fuel injection amount GFFST is
It is sufficient that the fuel-fuel mixture ignitable by the spark plug 18 is formed in a minimum amount, and the fuel consumption is not hindered by the ignition fuel injection, but rather is improved as described later.

Then, the ignition mixture of the fuel for ignition in the first fuel injection is used as an ignition source, and the flame is propagated to the fuel spray mixture of the main injection fuel for the second fuel injection to perform stratified combustion.

Therefore, as the second fuel injection, the main fuel injection for obtaining a predetermined output according to the engine operating state can be performed immediately before ignition, and the combustion temperature is reduced due to the presence of the fuel droplets. It is possible to reduce the generation of NOx. In addition, since the flame propagates to the air-fuel mixture formed by the second injected fuel due to the ignition of the air-fuel mixture by the first injected fuel, even in the lean air-fuel mixture as a whole in the combustion chamber 12, in the portion where combustion is performed, Therefore, it is possible to form an air-fuel mixture in the flammable range, and as a result, it is possible to greatly improve the lean limit and improve fuel efficiency.

Thus, smoldering can be prevented, and
By improving the ignitability and improving the air-fuel ratio lean limit, it is possible to achieve both improvement in fuel efficiency and improvement in exhaust emission.

On the other hand, in step S63 of the θ1 crank pulse interruption routine in FIG. 6, when the uniform mixed combustion of FSTR = 0 is selected, the process proceeds to step S66, where the first injection start timing timer for the fuel injection target cylinder is started. Exit the routine.

That is, when uniform mixed combustion is selected, as shown in FIG.
The first injection start timing timer of the corresponding cylinder is started by the input of the preceding θ1 crank pulse.

When uniform mixing combustion is selected, one cylinder 1
Since fuel injection is performed once per cycle, only the first injection start timing timer and the first fuel injection timer are used without using the second injection start timing timer and the second fuel injection timer. At this time, in the first injection start timing timer, the fuel injection start timing IJST which is set back from the compression top dead center of the corresponding cylinder by the fuel injection control routine is set. The injection start timing IJST is set on the basis of the compression top dead center of the corresponding cylinder to prevent fuel blow-through and to give a proper time for obtaining a uniform mixing state of fuel spray and fresh air at the time of ignition.

At this time, the first fuel injection timer includes a fuel injection pulse width Ti for obtaining a fuel injection amount for obtaining a predetermined engine output in accordance with the engine operating state at that time in accordance with the uniform mixed combustion. Is set.

When the fuel injection start timing IJST is reached by the timing of the first injection start timing timer, the above-described IJST1 interrupt routine of FIG. 7 is started, and in step S71, the first fuel injection timer of the corresponding cylinder is started. Start and exit the routine.

As a result, an injector drive signal based on the fuel injection pulse width Ti set in the first fuel injection timer is output to the injector 13 of the corresponding cylinder (see FIG. 14). Fuel is injected that adapts to a predetermined metered homogeneous mixed combustion corresponding to the fuel injection pulse width Ti and obtains a fuel injection amount for securing a predetermined engine output according to the engine operating state at that time.

When the BTDC θ1 crank pulse is input to the corresponding cylinder, the ignition timing timer and the energization start timing timer for the relevant cylinder are started.

When the energization start timing timer reaches the energization start timing TDWL, the above-described TDWL interrupt routine of FIG. 9 is started.
An energization signal for the corresponding cylinder is output to 0 (see FIG. 14), and energization (dwell) of the ignition coil 19 of the relevant cylinder is started.

Thereafter, when the timing of the ignition timing timer reaches the ignition timing TADV adapted to the uniform mixed combustion set in the ignition timing timer, the TADV interruption routine of FIG. 10 is started, and in step S101, the corresponding cylinder is started. The dwell for the ignition coil 19 is cut. As a result, a high secondary voltage is induced in the ignition coil 19 of the relevant cylinder, and the discharge electrode 18a of the spark plug 18 of the relevant cylinder sparks.

Here, the fuel injection timing IJST at the time of selecting the uniform mixing combustion is set to be calculated backward with reference to the compression top dead center of the corresponding cylinder, so that a uniform mixing state of fuel spray and fresh air can be obtained at the time of ignition. Fuel injection is started at an appropriate time.

Accordingly, in the state where the injected fuel and fresh air are sufficiently mixed in the combustion chamber 12, that is, under the uniform mixing state in which the fuel spray is sufficiently diffused, ignition is performed to ignite, and this uniform mixing is performed. The mixture in the state burns immediately. Thereby, at the time of high rotation and high load, a high average effective pressure is obtained by uniform mixed combustion, and the required output is secured.

In the present embodiment, the throttle opening α is used as an example of the engine load. However, the present invention is not limited to this. Instead of the opening degree α, an intake air amount, an intake pipe pressure downstream of the throttle valve, an intake air amount per intake stroke, or the like may be adopted.

In this embodiment, the ignition timing and each fuel injection timing are controlled by the time control method. However, the present invention is not limited to this, and the ignition timing and the fuel injection timing and the fuel injection timing are controlled by the angle control method. Of course, each fuel injection timing may be controlled.

[0164]

As described above, according to the first aspect of the present invention, the fuel injection is performed twice per one cycle of the cylinder, and as the first fuel injection, the air-fuel mixture ignitable by the ignition plug is minimized. The ignition fuel injection amount that can be formed is limited based on the operating state, and after the fuel of this ignition fuel injection amount is injected and the injected fuel collides with the piston top surface, the air-fuel mixture of the injected fuel is used as the ignition plug. Is calculated backward with reference to the ignition timing, and the time before the ignition or the crank angle is set as the ignition fuel injection timing, so that the fuel injection direction by the injector is directly directed to the ignition plug. The fuel spray from the first ignition fuel injection from the injector collides with the top surface of the piston, and synergizes with the in-cylinder intake air flow. When the air-fuel ratio mixture of coming combustible range rebounding of reaches the ignition plug, can be ignited by the spark plug to the mixture. Therefore, adhesion of fuel droplets to the discharge electrode of the ignition plug is suppressed, carbon is prevented from being deposited on the discharge electrode, and smoldering is prevented. At the position of the discharge electrode of the ignition plug, and the ignitability can be improved.

As the second fuel injection, a main fuel injection amount for obtaining a predetermined output according to the operating state is set, and the main fuel injection amount is determined after the ignition fuel injection based on the operating state. Since the main fuel injection timing for injecting fuel is set, main fuel injection for obtaining a predetermined output can be performed immediately before ignition, and the presence of fuel droplets lowers the combustion temperature and reduces NOx generation. can do. Further, since the flame propagates to the mixture formed by the second injected fuel due to the ignition of the mixture by the first injected fuel, even in the lean mixture as a whole in the combustion chamber, the flammable portion is not flammable. It is possible to form an air-fuel mixture in the range, and as a result,
The lean limit can be improved, and the fuel efficiency can be improved.

Therefore, smoldering can be prevented, ignitability can be improved, and fuel efficiency can be improved by improving the air-fuel ratio lean limit.
It is possible to achieve both improvement in exhaust emission.

According to the second aspect of the present invention, stratified charge combustion is selected at low engine speed and low load, and uniform mixed combustion is selected at high engine speed and high load based on the operating state of the engine. When stratified charge combustion is selected, two cycles per one cylinder
The first fuel injection is performed, and as the first fuel injection, an ignition fuel injection amount is set so that an air-fuel mixture ignitable by the spark plug can be formed at a minimum, and the fuel having the ignition fuel injection amount is injected before injection. After the fuel collides with the piston top surface, the time or crank angle until the mixture of the injected fuel reaches the spark plug is calculated back with reference to the ignition timing, and the time or crank angle before the ignition is used as the fuel injection for ignition. Since it is set as the timing, the fuel injection direction by the injector is arranged without directing to the ignition plug, and when stratified combustion is selected, the fuel spray from the first ignition fuel injection from the injector collides with the piston top surface. By the synergistic action with the in-cylinder intake air flow, when the rebounded air-fuel ratio mixture in the combustible range reaches the spark plug, the air-fuel mixture is reliably generated. It can be ignited by the spark plug. Accordingly, adhesion of fuel droplets to the discharge electrode of the ignition plug is suppressed, carbon is prevented from being deposited on the discharge electrode of the ignition plug, and while smoldering is prevented, the air-fuel mixture in the flammable range is ignited during ignition. Can reliably be present at the position of the discharge electrode of the ignition plug, and even during stratified combustion, stable ignition can be obtained, and ignitability can be improved.

As the second fuel injection, a main fuel injection amount adapted to stratified combustion is set in accordance with the engine operating condition, and the fuel based on the main fuel injection amount adapted to stratified combustion is set based on the engine operating condition. Since the main fuel injection timing for injecting NOx is set, the main fuel injection for obtaining a predetermined output in accordance with the engine operating state can be performed immediately before ignition, and the presence of fuel droplets lowers the combustion temperature to reduce NOx. Can be reduced. Also, 1
The ignition mixture of the second fuel injection fuel becomes the ignition type, and the flame propagates to the fuel spray mixture of the main injection fuel of the second fuel injection to perform stratified combustion. As for air, it is possible to form an air-fuel mixture in the flammable range in the part where combustion is performed, and as a result, the lean limit during stratified combustion can be significantly improved, further improving fuel efficiency can do.

On the other hand, when uniform mixed combustion is selected, fuel injection is performed once per one cylinder cycle, and a fuel injection amount suitable for uniform mixed combustion is set based on the engine operating condition, and the above-mentioned fuel injection amount adapted for uniform mixed combustion is set. Since the fuel injection timing for injecting the fuel based on the fuel injection amount is set according to the engine operating state, the ignition is performed to ignite under the uniform mixing state in which the fuel spray is sufficiently diffused in the combustion chamber, and this uniform mixing is performed. The mixture in the state burns immediately. As a result, at the time of high rotation and high load, a high average effective pressure is obtained by uniform mixed combustion, and the required output can be secured.

[Brief description of the drawings]

FIG. 1 is a basic configuration diagram of the present invention.

FIG. 2 is a flowchart of a cylinder discrimination / engine speed calculation routine.

FIG. 3 is a flowchart of an ignition control routine.

FIG. 4 is a flowchart of a fuel injection control routine.

FIG. 5 is a flowchart of a fuel injection control routine (continued).

FIG. 6 is a flowchart of a θ1 crank pulse interrupt routine.

FIG. 7 is a flowchart of an IJST1 interrupt routine.

FIG. 8 is a flowchart of an IJST2 interrupt routine.

FIG. 9 is a flowchart of a TDWL interrupt routine.

FIG. 10 is a flowchart of a TADV interrupt routine.

FIG. 11 is an explanatory diagram of an area determination value table.

FIG. 12 is a time chart showing a relationship among a crank pulse, a cam pulse, an ignition signal during stratified combustion, and an injector drive signal.

FIG. 13 is a time chart showing the relationship between ignition fuel injection, main fuel injection, and ignition in the same cylinder during stratified charge combustion;

FIG. 14 is a time chart showing a relationship between a crank pulse, a cam pulse, an ignition signal and an injector drive signal during uniform mixed combustion.

FIG. 15 is an overall schematic view of an in-cylinder direct injection engine.

FIG. 16 is an explanatory diagram showing a combustion chamber structure of the in-cylinder direct injection engine and showing a state at the time of fuel injection for ignition.

FIG. 17 is a bottom view of the cylinder head viewed from the combustion chamber side.

FIG. 18 is an explanatory diagram corresponding to FIG. 16 and showing a state at the time of main fuel injection.

FIG. 19 is an explanatory view showing a state at the time of ignition;

FIG. 20 is a front view of a crank rotor and a crank angle sensor.

FIG. 21 is a front view of a cam rotor and a cam angle sensor.

FIG. 22 is a circuit configuration diagram of an electronic control system.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 In-cylinder direct injection engine 11a Piston top surface 13 Injector 18 Spark plug 42 Crank angle sensor 50 Electronic control unit (ignition fuel injection amount setting means, main fuel injection amount setting means, ignition fuel injection timing setting means, main fuel injection Timing setting means, combustion method selection means, fuel injection amount setting means, fuel injection timing setting means) NE engine speed α throttle opening GFFST ignition fuel injection amount GFMAIN main fuel injection amount TADV ignition timing (ignition timing) IJSTFST ignition Fuel injection start timing (ignition fuel injection timing) IJSTMAIN Main fuel injection start timing (main fuel injection timing) GF Fuel injection amount IJST Fuel injection start timing (fuel injection timing)

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code FI F02D 41/34 F02D 41/34 HE

Claims (2)

[Claims] An in-cylinder direct injection engine fuel injection control system for directly injecting fuel into a cylinder and igniting and burning the injected fuel by means of a spark plug, wherein at least a mixture ignitable by a spark plug is formed. Ignition fuel injection amount setting means for setting an ignition fuel injection amount to be obtained based on an operation state; main fuel injection amount setting means for setting a main fuel injection amount for obtaining a predetermined output according to an operation state; The time or crank angle from when the fuel is injected for the fuel injection amount to when the mixture of the injected fuel reaches the spark plug after the injected fuel collides with the piston top surface is calculated back with reference to the ignition timing. An ignition fuel injection timing setting means for setting a time before ignition or a crank angle as an ignition fuel injection timing; and injecting fuel according to the main fuel injection amount after the ignition fuel injection. A main fuel injection timing setting means for setting a main fuel injection timing based on an operation state. 2. A fuel injection control for an engine which directly injects fuel into a cylinder and ignites the injected fuel by an ignition plug to burn and selectively performs stratified combustion and uniform mixed combustion in accordance with an engine operating state. A combustion system selection means for selecting stratified combustion at low engine speed and low load, and selecting uniform mixed combustion at high engine speed and high load, and an air-fuel mixture ignitable by a spark plug when stratified combustion is selected. Ignition fuel injection amount setting means for setting an ignition fuel injection amount capable of forming the minimum amount of fuel; main fuel injection amount setting means for setting a main fuel injection amount adapted to stratified combustion according to an engine operating state; The time or crank angle from the time when the fuel injected by the fuel injection amount is injected to the time the mixture of the injected fuel reaches the spark plug after the injected fuel collides with the piston top surface is determined. An ignition fuel injection timing setting means for calculating a time or a crank angle before the ignition as an ignition fuel injection timing based on the ignition timing as a reference, and injecting fuel according to the main fuel injection amount adapted to stratified combustion. Main fuel injection timing setting means for setting the main fuel injection timing based on the engine operating state; and fuel injection quantity setting means for setting a fuel injection amount adapted to the uniform mixed combustion based on the engine operating state when uniform mixed combustion is selected. And fuel injection timing setting means for setting a fuel injection timing for injecting fuel with the fuel injection amount adapted to uniform mixed combustion based on an engine operating state. Control device.