JP2002332886A - Fuel injection controller of internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve combustion state of an engine by effectively utilizing fuel vapor supplied to the engine through purging. SOLUTION: Fuel vapor is introduced from an evaporation fuel purge device into an intake passage 10 of a cylinder fuel injection engine 1. An intake oxygen density sensor 31 is installed in the intake passage, to detect fuel vapor amount in the intake. An ECU 30 corrects fuel injection amount in the cylinder fuel injection nozzles 111 to 114, according to the fuel vapor amount detected, so that the air-fuel ratio of the engine when purging is conducted becomes leaner than that when the purging is not conducted, and changes the fuel injection start timing and the fuel injection end timing, according to the amount of fuel vapor. Thus, operation at leaner air-fuel ratio can be performed, while maintaining engine output, when the purging is performed; and reduction in fuel cost and improvement in exhaust properties can be achieved.

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 an internal combustion engine, and more particularly to a fuel injection control device for an engine having an in-cylinder fuel injection valve for directly injecting fuel into a cylinder.

[0002]

2. Description of the Related Art Evaporated fuel (fuel vapor) in a fuel tank is once adsorbed by a canister containing activated carbon or the like, and the fuel vapor adsorbed by the activated carbon is supplied (purged) to an engine intake passage during operation of the engine. 2. Description of the Related Art An evaporative fuel purging device that prevents fuel vapor in a tank from being released into the atmosphere by burning the fuel in a tank is generally known.

When purging the fuel vapor, an excessive amount of fuel corresponding to the fuel vapor amount is supplied to the engine together with the intake air. Therefore, the fuel injection amount to the engine is set to the same value as when the purge is not executed. , The air-fuel ratio of the engine changes (decreases), and the combustion state of the engine may deteriorate. For this reason, the amount of fuel injection to the engine is reduced and corrected by the amount corresponding to the supplied fuel vapor when purging is performed, compared to when purging is not performed.

An example of an engine that performs this type of weight loss correction is disclosed in, for example, Japanese Patent Application Laid-Open No. 2000-27716. The engine of the publication is a spark ignition engine provided with an in-cylinder fuel injection valve that injects fuel directly into a cylinder, injects fuel in a compression stroke of each cylinder, and is compressed in a cylinder and does not include fuel. The engine is capable of performing a so-called stratified combustion that forms a mixture layer having a combustible air-fuel ratio only in the vicinity of a spark plug in the air.

In an engine that performs stratified charge combustion, when purging is performed during stratified charge combustion, air containing fuel vapor is supplied into the cylinder, so that fuel injected from the fuel injection valve is further supplied during the compression stroke. In this case, the air-fuel ratio of the combustible air-fuel mixture formed decreases and the combustion deteriorates due to excessive enrichment of the air-fuel ratio of the combustible air-fuel mixture. Sometimes there was a problem that purging could not be performed.

In the engine disclosed in Japanese Patent Application Laid-Open No. 2000-27716, a swirling flow of intake air is formed in a cylinder during an intake stroke.
The fuel vapor is supplied into the cylinder along with the swirling flow so that the fuel vapor exists only in one of the air layer and the combustible air-fuel mixture layer in the cylinder, and the fuel paper amount is increased. By reducing the fuel injection amount by an amount corresponding to the above, the problem of the disturbance of the stratified combustion is reduced.

That is, since the fuel injection amount is reduced by the amount of fuel supplied as the vapor, the sum of the fuel injection amount and the vapor amount supplied to the cylinder is the same as when the purge is not performed even when the purge is executed. Amount will be maintained. This prevents the problem of stratified combustion turbulence due to the enrichment of the mixture air-fuel ratio during stratified combustion, and maintains the air-fuel ratio constant regardless of the presence or absence of purge throughout the cylinder. Of the exhaust gas due to the change (richness) of the exhaust gas is prevented.

However, Japanese Patent Application Laid-Open No. 2000-27716 discloses
In the engine disclosed in Japanese Patent Application Laid-Open Publication No. H10-209, although the fuel vapor is unevenly distributed in the cylinder and the fuel injection amount is corrected to be reduced according to the amount of the fuel vapor, there is a problem that the disturbance of the stratified combustion cannot be completely prevented.

For example, in an in-cylinder fuel injection engine, the fuel injection timing (that is, the fuel injection start and end timings) greatly affects the state of formation of the air-fuel mixture. For this reason, the fuel injection timing is set precisely so as to obtain an optimal air-fuel mixture according to the fuel injection amount, the engine speed, the load, and the like. Therefore, when the fuel injection amount changes, the optimum fuel injection timing changes even in other operating conditions, for example, the same engine speed and load. Since the fuel injection amount is usually changed by changing the valve opening time (injection time) of the fuel injection valve, even in the engine disclosed in the above publication, the fuel injection timing temporarily changes when the fuel injection amount changes. However, the fuel injection time is usually changed by fixing either the opening timing of the fuel injection valve (fuel injection start timing) or the closing timing of the fuel injection valve (fuel injection end timing) and changing the other. Control is performed. For this reason, in the device of the above publication, even when the fuel injection amount is corrected to be reduced, the fuel injection start timing or the fuel injection end timing is fixed, and optimal fuel injection cannot be performed in view of the fuel injection amount and the engine operating state. There's a problem.

In order to solve the above-mentioned problem, the applicant of the present application has already disclosed in Japanese Patent Application No. 2000-278681 the start timing and the end timing of fuel injection from each cylinder in accordance with the amount of fuel vapor at the time of purging. A fuel injection control device in which both are changed has been proposed.

That is, in the apparatus of the above-mentioned application, the amount of fuel is reduced by an amount corresponding to the amount of fuel vapor at the time of purging,
The combustion air-fuel ratio in the cylinder is corrected so as not to change depending on the presence or absence of the purge, but at the same time, both the start time and the end time of the fuel injection are changed according to the fuel vapor amount, and the fuel supplied to the cylinder by the fuel injection is changed. And the fuel supplied to the cylinder as fuel vapor forms an optimal air-fuel mixture. As a result, both the fluctuation of the combustion air-fuel ratio and the deterioration of the combustion during the execution of the purge can be prevented.

[0012]

However, the apparatus disclosed in Japanese Patent Application Laid-Open No. 2000-27718 considers fuel vapor supplied by purging only as a disturbance to air-fuel ratio control. For this reason, the device of the above publication controls the fuel injection amount by focusing only on preventing the deterioration of combustion when the fuel vapor is supplied to the cylinder.

However, in general, the fuel injected into the cylinder does not generate a completely uniform mixture in the cylinder, and a fuel concentration distribution (concentration "unevenness") occurs in the mixture.
In particular, when the engine is operated at an air-fuel ratio (lean air-fuel ratio) higher than the stoichiometric air-fuel ratio, the above-mentioned unevenness of the concentration in the air-fuel mixture is a major cause of deterioration of combustion. On the other hand, the fuel supplied to the cylinder as a fuel vapor during the execution of the purge forms an extremely uniform mixture. Therefore, if the fuel vapor is effectively used, combustion in the cylinder should be able to be improved. However, Japanese Patent Laid-Open No. 2000-27
Both JP-A-718 and JP-A-2000-278681 regard the fuel vapor at the time of purging as only a disturbance to the air-fuel ratio control, and do not consider how to effectively utilize it.

In view of the foregoing, it is an object of the present invention to improve the combustion in a cylinder by effectively utilizing fuel vapor supplied to the cylinder at the time of performing a purge.

[0015]

According to the first aspect of the present invention, there is provided an evaporative fuel purge device for supplying evaporative fuel in a fuel tank to an engine intake passage, and detecting the concentration of the evaporative fuel in engine intake air. And a fuel injection setting unit that sets a fuel injection amount from the in-cylinder fuel injection valve based on an engine operating state. A fuel injection control device for an internal combustion engine, further comprising an engine combustion chamber based on a vaporized fuel concentration detected by the vapor detecting means when the vaporized fuel is supplied from the vaporized fuel purge device to an engine intake passage. So that the total of the amount of evaporated fuel supplied to the engine and the amount of fuel supplied to the engine combustion chamber by fuel injection is smaller than the fuel injection amount set by the fuel injection setting means. And the fuel injection amount correcting means for correcting the constant fuel injection amount, the fuel injection control apparatus for an internal combustion engine having a are provided.

That is, according to the first aspect of the present invention, the fuel injection amount of the engine is reduced to a value equal to or larger than the fuel amount supplied to the engine as fuel vapor during the execution of the purge. For this reason, the air-fuel ratio of the engine is leaner when purging is performed than when purging is not performed, even under the same operating conditions (rotational speed, load, etc.). The fuel vapor supplied to the engine by the purge forms a uniform mixture with air. For this reason, during the execution of the purge, the unevenness of the concentration of the air-fuel mixture in the cylinder is reduced, and particularly during the lean air-fuel ratio operation, the flame propagation during the combustion is improved, and the combustion state is improved. For this reason, the lean limit of combustion becomes high when purging is performed,
Operation at a leaner air-fuel ratio becomes possible. Further, as a result of the improved combustion, the engine output increases, so that the same engine output can be obtained even when the engine is operated at a leaner air-fuel ratio. According to the first aspect of the present invention, the engine air-fuel ratio is shifted during the execution of the purge by effectively utilizing the uniform air-fuel mixture supplied by the purge. This improves the engine fuel economy, so emissions of the NO _X by the combustion are reduced.

According to the second aspect of the present invention, the fuel injection amount correcting means sets the difference between the fuel injection amount set by the setting means and the total fuel amount as the detected fuel vapor concentration increases. 2. The fuel injection control device according to claim 1, wherein the fuel injection amount set by the setting means is corrected so that the fuel injection amount becomes larger.

That is, in the second aspect of the present invention, the engine air-fuel ratio at the time of executing the purge is shifted to the lean side as the concentration of the evaporated fuel in the intake air increases, as compared with the case where the purge is not executed. Since the vaporized fuel supplied by the purge is a uniform mixture, the larger the amount of fuel supplied as the vaporized fuel, the more uniform the mixture in the cylinder, and the less the fuel concentration becomes uneven. For this reason, the leaner the combustion becomes, the leaner the fuel vapor concentration becomes at the time of performing the purge, so that the air-fuel ratio can be largely shifted to the leaner side.

According to the invention of claim 3, according to claim 1,
The fuel injection control device according to any one of claims 1 to 3, wherein the fuel injection control device performs fuel injection during an intake stroke of each cylinder from the fuel injection valve and performs mode fuel injection for forming a uniform mixture in the cylinder.

According to a fourth aspect of the present invention, there is provided the fuel injection control device according to the first aspect, wherein the fuel is injected from the fuel injection valve in the intake stroke of each cylinder to uniformly mix the cylinder. A fuel injection control device that performs fuel injection in a weak stratified combustion mode, in which fuel is injected during a compression stroke of each cylinder while forming a fuel mixture to form a layer of a mixture having a low air-fuel ratio in the uniform mixture. You.

According to a fifth aspect of the present invention, there is provided the fuel injection control device according to the first aspect, wherein fuel is injected from the fuel injection valve in a compression stroke of each cylinder, and the fuel in the air in the cylinder is released. A fuel injection control device for performing a stratified combustion mode fuel injection for forming a layer of a combustible air-fuel mixture is provided.

According to a sixth aspect of the present invention, there is provided the fuel injection control device according to the first aspect, wherein the fuel is injected from the fuel injection valve to the intake stroke of each cylinder in accordance with engine operating conditions. And a uniform combustion mode fuel injection for forming a uniform mixture in the cylinder, and a fuel injection from the fuel injection valve in an intake stroke of each cylinder to form a uniform mixture in the cylinder and a compression stroke for each cylinder. A weak stratified combustion mode fuel injection, which forms a layer of a mixture having a low air-fuel ratio in the uniform mixture, and performs a fuel injection from the fuel injection valve in a compression stroke of each cylinder. There is provided a fuel injection control device that performs fuel injection by selecting one of stratified combustion mode fuel injection that forms a layer of a combustible air-fuel mixture in air.

In other words, according to the third to sixth aspects of the present invention, the correction of the fuel injection amount according to the evaporated fuel concentration according to the first aspect is performed when the homogeneous mixture combustion operation is performed or when the weak stratified combustion operation is performed. It is applied to either the case where the stratified combustion operation is performed or the case where these operation modes are switched according to the engine operation state. Therefore, in the inventions of claims 3 to 6, the fuel injection correction of claim 1 is applied to all of the operation modes of the in-cylinder fuel injection type spark ignition engine. For this reason, even in any operating state of the in-cylinder fuel injection type spark ignition engine, it is possible to shift the engine air-fuel ratio to the lean air-fuel ratio side by effectively using the evaporated fuel in the intake air by the purge. at the same time it is possible to achieve a reduction in the reduction and NO _X emissions fuel consumption in.

According to the seventh aspect of the present invention, there is further provided a fuel injection timing setting means for setting a fuel injection timing from the in-cylinder fuel injection valve based on an engine operation state, and the fuel injection control means The fuel injection control device according to claim 1, wherein both the fuel injection start time and the fuel injection end time set by the fuel injection time setting means are changed according to the detected fuel vapor concentration.

That is, according to the present invention, when the fuel injection amount is reduced and corrected to shift the engine air-fuel ratio to the lean side based on the detected fuel vapor concentration, the fuel injection is performed in accordance with the fuel vapor concentration. Since both the start time and the end time are changed, the air-fuel mixture formed in the cylinder is further optimized for the combustion state, and the combustion state is further improved.

[0026]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a diagram showing a schematic configuration of an embodiment when the present invention is applied to an internal combustion engine for a vehicle. In FIG. 1, reference numeral 1 denotes an automobile internal combustion engine. In the present embodiment, the engine 1 is a four-cylinder gasoline engine having four cylinders # 1 to # 4, and each cylinder is provided with in-cylinder fuel injection valves 111 to 114 for directly injecting fuel into the cylinder. ing. As will be described later, the internal combustion engine 1 of the present embodiment is an engine capable of operating in a wide range of air-fuel ratios from an air-fuel ratio higher than the stoichiometric air-fuel ratio (lean) to an air-fuel ratio lower than the stoichiometric air-fuel ratio (rich). I have.

Further, in this embodiment, the cylinders # 1 to # 4 are grouped into two cylinder groups including two cylinders whose ignition timings are not continuous with each other. (For example, in the embodiment of FIG. 1, the cylinder ignition order is 1-3-4-2,
The cylinders # 1 and # 4 and the cylinders # 2 and # 3 each constitute a cylinder group. The exhaust port of each cylinder is connected to an exhaust manifold for each cylinder group, and is connected to an exhaust passage for each cylinder group. In FIG. 1, 21a is # 1, # 4
The exhaust manifold 21b connects the exhaust ports of the cylinder group consisting of cylinders to the individual exhaust passages 2a, and the exhaust manifold 21b connects the exhaust ports of the cylinder group consisting of # 2 and # 4 cylinders to the individual exhaust passage 2b. In the present embodiment, start catalysts (hereinafter, referred to as "SC") 5a and 5b made of a known three-way catalyst are arranged on the individual exhaust passages 2a and 2b, respectively. The individual exhaust passages 2a and 2b are S
It joins the common exhaust passage 2 on the downstream side of C.

Reference numerals 29a and 29b in FIG. 1 denote air-fuel ratio sensors disposed upstream of the start catalysts 5a and 5b in the individual exhaust passages 2a and 2b. Air-fuel ratio sensor 2
Reference numerals 9a and 29b denote sensors for outputting a voltage signal corresponding to the exhaust air-fuel ratio in a wide air-fuel ratio range, and the output is used for air-fuel ratio control of the engine 1.

In FIG. 1, reference numeral 10b denotes an intake manifold connecting each intake port of each cylinder of the engine to an intake passage 10;
a is a surge tank provided in the intake passage 10. Further, in this embodiment, the throttle valve 1 is provided on the intake passage 10.
5 are provided. The throttle valve 15 of the present embodiment is a so-called electronically-controlled throttle valve, and is driven by an appropriate type of actuator 15a such as a stepper motor or the like, and has an opening in accordance with a control signal from an ECU 30 described later.

A known fuel vapor purge device 40 is connected to the intake passage 10 downstream of the throttle valve 15 via a purge control valve 41. The purge device 40 includes a canister containing an adsorbent such as activated carbon, for example.
The fuel vapor in the fuel tank (not shown) of the engine 1 is adsorbed by the adsorbent in the canister. This prevents the fuel vapor from escaping from the fuel tank to the atmosphere. The purge control valve 41 includes an actuator of an appropriate type such as a stepper motor, and takes an opening degree according to a control signal of the ECU 30. When the purge control valve 41 is opened during operation of the engine, the fuel vapor adsorbed in the canister of the purge device 40 flows into the intake passage 10 from the purge control valve 41, and the engine vapor passes through the throttle valve 15. The mixture is mixed to form a uniform mixture and is sucked into each cylinder of the engine 1.

Further, in this embodiment, an oxygen concentration sensor 31 for detecting the oxygen concentration in the intake air is disposed in the surge tank 10a of the intake passage 10. Oxygen concentration sensor 3
1 outputs a signal corresponding to the oxygen concentration in the intake air in a wide concentration range, and the same type as the air-fuel ratio sensors 29a and 29b is used. Reference numeral 30 in FIG. 1 denotes an electronic control unit (ECU) of the engine 1. ECU
Reference numeral 30 denotes a microcomputer having a known configuration including a RAM, a ROM, and a CPU in the present embodiment, and performs basic control such as ignition timing control and air-fuel ratio control of the engine 1. In the present embodiment, in addition to performing the above basic control, the ECU 30 changes the fuel injection mode of the in-cylinder injection valves 111 to 114 according to the engine operating state to change the operating air-fuel ratio of the engine, as described later. In addition to the control, the opening and closing of the purge control valve 41 is controlled to purge the fuel vapor. Further, the ECU 30 detects the amount of fuel vapor during intake by using the intake oxygen concentration sensor 31 during execution of the purge, and based on the amount of fuel vapor, the fuel injection amount of the fuel injection valves 111 to 114 of each cylinder, the fuel injection amount, Purge fuel injection control for changing the timing and the like is performed.

In order to perform each of the above controls, an input port of the ECU 30 is provided with signals representing the exhaust air-fuel ratio at the inlet of the start catalysts 5a, 5b from the air-fuel ratio sensors 29a, 29b, and the oxygen concentration in the intake oxygen sensor 31. A signal representing the concentration and a signal corresponding to the intake pipe pressure of the engine are input from an intake pressure sensor 33 provided in the engine intake manifold. An input port of the ECU 30 is provided with a rotation angle pulse signal indicating a crank rotation angle at every constant rotation angle of the engine crankshaft (for example, every 15 degrees) from a crank angle sensor 33 arranged near the engine crankshaft (not shown). And a reference crank position pulse signal indicating a crank shaft reference position at every 720 degrees of engine rotation (for example, at each compression top dead center of the # 1 cylinder).

Further, in this embodiment, the input port of the ECU 30 indicates the driver's accelerator pedal depression amount (accelerator opening amount) from an accelerator opening sensor 37 arranged near an accelerator pedal (not shown) of the engine 1. Signal is input. The ECU 30 controls the intake pressure sensor 33 at predetermined intervals.
The ECU 30 converts the output from the output of the accelerator opening sensor 37 into an A / D signal to obtain an intake pipe pressure PM and an accelerator opening ACCP.
Is stored in a predetermined area of the RAM. Also, the ECU 30
Are the engine speed NE based on the interval between the rotation angle pulse signals input from the crank angle sensor 33 at regular intervals, and the crank angle based on the number of rotation angle pulse signals after the reference crank position pulse signal is input. (Transfer) is calculated and used for various controls.

The output port of the ECU 30 is connected to the fuel injection valves 111 to 114 of the respective cylinders via a fuel injection circuit (not shown) in order to control the amount and timing of fuel injection into each cylinder. The throttle valve 15 is connected to an actuator 15b of the throttle valve 15 via a drive circuit (not shown) to control the opening of the throttle valve 15. Also, ECU
Reference numeral 30 is connected to an actuator of the purge control valve 41 via a drive circuit (not shown) to control the opening degree of the purge control valve 41 to purge the fuel vapor.

In this embodiment, during operation of the engine 1, the ECU 30 operates the engine 1 in one of the following five modes according to the operating conditions. (a) Lean air-fuel ratio stratified combustion (one compression stroke injection) (b) Lean air-fuel ratio weakly stratified combustion (one injection each for intake stroke / compression stroke) (c) Lean air-fuel ratio uniform mixture combustion (one intake stroke) (D) Combustion of stoichiometric air-fuel ratio uniform mixture (injection once in intake stroke) (e) Combustion of rich air-fuel ratio uniform mixture (injection stroke once)

That is, in the light load operation region of the engine 1,
The lean air-fuel ratio stratified combustion in the mode (a) is performed.
The engine 1 is provided with two intake valves, namely, an intake valve having a swirl port for generating a swirl (swirl flow) of intake air in a cylinder, and an intake valve having a normal straight port. By adjusting the opening of a swirl control valve (SCV) (not shown) provided in the passage, it is possible to control the amount of intake air flowing into the cylinder from the swirl port. When performing stratified charge combustion, the SCV opening is fully closed to increase the amount of intake air from the swirl port and generate strong swirl in the cylinder. In this state, in-cylinder fuel injection is performed only once during the compression stroke of each cylinder, and the injected fuel forms a combustible air-fuel mixture layer near the cylinder ignition plug. Further, the fuel injection amount in this operating state is extremely small, and the air-fuel ratio as a whole in the cylinder becomes about 25 to 30 or more.

When the load increases from the state of the mode (a) to the low load operation region, the lean air-fuel ratio weak stratified combustion in the mode (b) is performed. As the engine load increases, the amount of fuel injected into the cylinder increases.However, in this load region, in addition to the fuel injection in the latter half of the compression stroke, the target amount of fuel is previously injected into the cylinder by injecting fuel in the first half of the intake stroke. I am trying to supply. The fuel injected into the cylinder in the first half of the intake stroke produces an extremely lean homogeneous mixture by the time of ignition. In the latter half of the compression stroke, fuel is further injected into this extremely lean homogeneous mixture, and a layer of ignitable combustible mixture is generated near the ignition plug. At the time of ignition, the combustible air-fuel mixture layer starts burning, and the flame propagates to the surrounding lean air-fuel mixture layer, so that stable combustion is performed.
In this state, the fuel amount supplied by the injection in the intake stroke and the compression stroke is increased from the mode (a), but the air-fuel ratio as a whole is moderately lean (for example, 20 to 3 in the air-fuel ratio).
0).

When the engine load further increases, the engine 1 performs the lean air-fuel ratio uniform mixture combustion in the mode (c). In this state, the SCV is fully opened, and most of the intake air flows into the cylinder from the straight port. In this state, the fuel injection is performed only once in the first half of the intake stroke, and the fuel injection amount is further increased from the mode (b). In this state, the homogeneous mixture generated in the cylinder has a lean air-fuel ratio relatively close to the stoichiometric air-fuel ratio (for example, an air-fuel ratio of about 15 to 25).
Becomes

When the engine load further increases and the engine enters a high-load operation range, the amount of fuel is further increased from the state of the mode (c), and the stoichiometric air-fuel ratio uniform mixture operation in the mode (d) is performed. In this state, a homogeneous air-fuel mixture having a stoichiometric air-fuel ratio is generated in the cylinder, and the engine output increases.
Further, when the engine load further increases and the engine becomes full load operation, the fuel injection amount is further increased from the state of the mode (d), and the rich air-fuel ratio uniform mixture operation in the mode (e) is performed.
In this state, the air-fuel ratio of the homogeneous mixture generated in the cylinder becomes rich (for example, the air-fuel ratio is about 12 to 14).

In this embodiment, the optimum operation mode (the above (a) to (e)) is determined in advance based on experiments and the like in accordance with the accelerator opening (the amount of depression of the accelerator pedal by the driver) and the engine speed.
Are stored in the ROM of the ECU 30 as a numerical table (map) using the accelerator opening and the engine speed. During the operation of the engine 1, the ECU 30 determines which of the above (a) to (e) is currently in operation, based on the accelerator opening detected by the accelerator opening sensor 37 and the engine speed, Control amounts for controlling the operating state of the engine, such as the fuel injection amount, fuel injection timing and number, ignition timing, throttle valve opening, and EGR amount (EGR valve opening) are determined according to each mode.

When the mode (d) (stoichiometric air-fuel ratio uniform mixture combustion) is selected, the ECU 30 further adjusts the fuel injection amount calculated above so that the engine exhaust air-fuel ratio becomes the stoichiometric air-fuel ratio. The air-fuel ratio control for feedback correction is performed based on the outputs of the air-fuel ratio sensors 29a and 29b. More specifically, when the modes (a) to (c) (lean air-fuel ratio combustion) are selected, the ECU 30 performs the operations based on the numerical tables prepared in advance for the modes (a) to (c). The control amounts such as the fuel injection amount, the fuel injection timing, the throttle opening, the EGR amount, and the ignition timing are determined from the accelerator opening and the engine speed. If the modes (d) and (e) (stoichiometric air-fuel ratio or rich air-fuel ratio uniform mixture combustion) are selected, the ECU 30 is prepared in advance for each of the modes (d) and (e). Based on the numerical value table, a control amount such as a fuel injection amount is set based on the throttle valve opening, the engine speed, and the intake pipe pressure detected by the intake pressure sensor 33.

The opening of the throttle valve 15 is set in the mode (a).
In (c), control is performed according to the accelerator opening in a region close to full open. In this region, when the accelerator opening decreases, the throttle valve opening also decreases.However, since this is a region equivalent to the throttle valve being fully opened, the intake pipe pressure becomes substantially constant even when the throttle valve opening changes, and the intake throttle is almost completely closed. Does not occur. On the other hand, in modes (d) and (e), the throttle valve opening is controlled to an opening substantially equal to the accelerator opening. That is, when the accelerator opening (accelerator pedal depression amount) is 0, the throttle opening is also 0 (fully closed), and when the accelerator opening is 100% (when the accelerator pedal is fully depressed), the throttle opening is Is also set to 100 percent (fully open). As described above, the form of fuel injection in the present embodiment is roughly divided into intake stroke fuel injection (modes (c), (d), and (e)) for forming a homogeneous mixture and compression stroke fuel for stratification. Injection (mode (a)). In the mode (b) weak stratified combustion, both of these are performed.

According to the study of the present inventor, it is possible to increase the lean limit at the time of executing the purge compared to the case where the purge is not performed in any of the homogeneous mixture combustion, the weak stratified combustion, and the stratified combustion. It turns out that there is. It is considered that the reason why the fuel injection amount can be reduced to be equal to or more than the amount of the fuel vapor supplied by the purge during the execution of the purge is as follows.

In the intake stroke fuel injection for forming a uniform mixture, it is preferable that the injected fuel forms a uniform mixture in the cylinder as much as possible. However, the fuel actually injected from the fuel injection valve is composed of fine droplet particles. The finer the fuel particle diameter, the easier it is to vaporize and a uniform mixture is formed. However, there is a limit to atomization of fuel in an actual fuel injection valve. In addition, in the in-cylinder injection valve, since the injected fuel is mixed with air in the cylinder for a short time, the dispersion of the fuel particles in the air-fuel mixture may not be uniform, and the vaporization may be insufficient.

Therefore, the mixture formed by the fuel injected by the fuel injection during the intake stroke fuel injection for forming a uniform mixture has a non-uniform fuel concentration distribution, that is, a mixture richer than the average air-fuel ratio of the mixture in the cylinder. There is a portion (over-rich region) and a portion leaner than the average air-fuel ratio (over-lean region). For this reason, if the average air-fuel ratio is made lean to some extent, in the over-lean portion, the propagation of the flame during combustion becomes insufficient, and the combustion deteriorates. Therefore, the lean limit during operation is determined by the air-fuel ratio in the over-lean region, and it is necessary to increase the entire fuel injection amount until the air-fuel ratio in the over-lean region can be set to a level that does not hinder the propagation of the flame. I will. As a result, in the over-rich region, the air-fuel ratio becomes richer than necessary, although the air-fuel ratio can be originally made leaner. On the other hand, the fuel vapor supplied from the purge device to the intake passage is in a vaporized state or in extremely fine particles even if it is not vaporized. Further, the fuel vapor sufficiently mixes with the intake air before flowing into the cylinder from the intake passage to form an extremely uniform mixture.

That is, in the intake stroke injection, as the amount of fuel vapor supplied as a uniform mixture increases, the fuel injection amount conventionally required to eliminate the over-lean region decreases, so that The fuel injection amount can be reduced. Therefore, as the amount of vapor during intake increases, the combustion state in the cylinder improves, and the lean limit increases as a whole.

In the compression stroke fuel injection for stratification, it is necessary to form a rich (rich) mixture layer around the spark plug without diffusing the injected fuel into the cylinder. Ideally, the injected fuel is concentrated only in this mixture, and the surroundings of the mixture are preferably air containing no fuel. However, it is actually difficult to completely separate the air-fuel mixture from the air, and part of the fuel injected to form the rich air-fuel mixture diffuses around the air-fuel mixture, resulting in a rich mixture. An air-fuel mixture layer having a lean air-fuel ratio (overlean region) is formed outside the gas layer.

If the air-fuel ratio is too lean in the over-lean region, the flame ignited in the rich mixture layer may not propagate to the over-lean region, and the fuel in the over-lean region may not burn. If the surroundings of the rich mixture layer are completely air only, there is no problem even if the flame does not propagate, but if the flame does not propagate to the over-lean area, HC and HC in the exhaust due to unburned fuel in the over-lean area Problems such as an increase in the amount of CO and a decrease in engine output occur. For this reason, even in the case of stratified charge combustion, it is necessary to set the fuel injection amount so that the air-fuel ratio of the over-lean layer formed around the rich mixture layer becomes an air-fuel ratio sufficient for flame propagation. There is a need to inject more fuel than quantity. That is, also in the stratified charge combustion, the fuel injection amount that can ensure the flame propagation in the over lean layer is the lean limit. On the other hand, when the fuel vapor having a uniform mixture is supplied to the cylinder by the purge, the fuel injected in the compression stroke is stratified in the uniform mixture. For this reason, the air-fuel ratio of the over-lean layer formed around the rich mixture layer becomes richer than when the purge is not performed. Therefore, it is not necessary to increase the amount of fuel in order to ensure the propagation of the flame in the overlean layer, and it is possible to reduce the fuel injection amount to the amount of fuel vapor supplied by the purge. In other words, the lean limit is expanded when purging is performed, compared to when purging is not performed.

As described above, in the case of weak stratified charge combustion in which the intake stroke fuel injection and the compression stroke injection are performed and a rich mixture is formed in the lean mixture formed by the intake stroke fuel injection, as described above. When the fuel vapor, which is a uniform mixture obtained by the purge, is supplied to the cylinder, the over-lean region in the lean mixture formed by the fuel injection during the intake stroke is eliminated, and the propagation of the flame becomes good. The injection amount can be reduced. For this reason, during the execution of the purge, the fuel injection amount can be reduced even in the case of weak stratified combustion, and the lean limit is expanded.

Next, the fuel injection control of this embodiment will be described. In the present embodiment, as will be described in detail later, the amount of fuel vapor during intake is calculated based on the intake oxygen concentration detected by the intake oxygen concentration sensor 31. Then, the fuel injection amount and the fuel injection timing (base fuel injection amount TAU, base fuel injection timing KINJT) when purging is not performed are determined based on the load and rotation speed of the engine, and the calculated base fuel injection amount is used as the fuel. Correct according to the amount of vapor.

Conventionally, the correction of the base fuel injection amount based on the fuel vapor amount is performed by subtracting the fuel amount corresponding to the fuel vapor from the base fuel injection amount, and the sum of the fuel vapor amount and the corrected fuel injection amount is calculated as the base fuel injection amount. It was set to be equal to the injection amount. On the other hand, in the present embodiment, the sum of the fuel vapor amount and the corrected fuel injection amount is set to be smaller than the base fuel injection amount.

In this embodiment, as will be described later, in this embodiment, a vapor equivalent value B corresponding to the ratio between the fuel vapor amount (concentration) in intake air and the base fuel injection amount TAU is calculated from the oxygen concentration sensor output. The correction coefficient (reduction coefficient) D of the fuel injection amount is calculated using a predetermined relationship based on the vapor equivalent value B. The actual fuel injection amount from the fuel injection valve is obtained by multiplying the base fuel injection amount TAU by the reduction coefficient D. Since the vapor equivalent value B is a value corresponding to the ratio between the fuel vapor amount during intake and the base fuel injection amount TAU, the sum of the fuel vapor amount and the corrected fuel injection amount is equal to the base fuel injection amount as in the related art. When it becomes equal to the injection amount, D + B = 1. However, in this embodiment, D +
The difference is that B = γ ≦ 1 is set. γ is a weight loss constant, and is a value determined from the vapor equivalent value B.

The value of the weight loss constant γ is set to a smaller value as the vapor equivalent value B is larger. The vapor equivalent value B is
This corresponds to the ratio between the fuel vapor amount during intake and the base fuel injection amount. As described above, if the base fuel injection amount is constant, the combustion of the engine is improved and the fuel injection amount can be reduced as the vapor amount increases. For this reason, the value of γ is set to a smaller value as the vapor equivalent value B is larger, and the weight loss coefficient D is smaller. Further, when the vapor concentration is smaller than the base fuel injection amount (that is, the vapor equivalent value B
Is small), the effect of the fuel vapor in the intake air on the combustion becomes relatively small, and the effect of improving the combustion by the fuel vapor becomes relatively small. Therefore, the value of γ is set to approach 1 as the value of the vapor equivalent value B decreases. The reduction coefficient D represents the proportion of fuel that can be reduced without reducing the engine output during execution of the purge.

The value of the reduction constant γ actually varies depending on the type of engine and the mode of fuel injection (intake stroke injection, compression stroke injection). Driving by changing the amount of fuel vapor
It is calculated by finding the amount of fuel that can be reduced without lowering the engine output by experiment for each combustion mode. As described above, in the present embodiment, during the execution of the purge, the fuel injection amount is equal to TAU × (1-D) with respect to the base fuel injection amount TAU.
And the fuel injection time is also reduced by a factor of (1-D). In this case, if either the fuel injection start time or the fuel injection end time is fixed, the injected fuel may not be able to form an appropriate air-fuel mixture. In some cases, a sufficient combustion improving effect cannot be obtained.

Therefore, in this embodiment, when shortening the fuel injection time, both the fuel injection start timing and the fuel injection end timing are changed according to the amount of vapor during intake. Next, the fuel injection timing control at the time of the fuel vapor purge according to the present embodiment will be described. In this specification, in order to distinguish the fuel injection amount and the fuel injection timing at the time of performing the purge,
The fuel injection amount and the fuel injection timing set by the ECU 30 when the purge is not being performed are called a base fuel injection amount and a base fuel injection timing, respectively.

As described above, the form of fuel injection in the present embodiment is roughly classified into two types, that is, fuel injection in the intake air amount stroke for forming a uniform mixture and fuel injection in the compression stroke for stratification.
Therefore, the conditions required for the intake stroke fuel injection and the compression stroke fuel injection will be described first.

First, in the fuel injection during the intake stroke, it is necessary to form a uniform mixture in the cylinder so that the injected fuel is diffused as uniformly as possible in the cylinder. For this purpose, it is necessary to allow sufficient time for the injected fuel to diffuse into the cylinder, so that the intake stroke fuel injection is performed as early as possible (that is, as close as possible to the intake stroke top dead center).
Preferably, it is completed. On the other hand, if the piston is at the ascending position during the execution of fuel injection, the injected fuel adheres to the piston, and formation of a uniform air-fuel mixture is hindered.
For this reason, it is desirable that the intake stroke fuel injection be started when the piston is at the lowered position as much as possible (that is, at a time as close as possible to the bottom dead center of the intake stroke).

That is, it is desirable that the intake stroke fuel injection be started as late as possible (closer to the bottom dead center) and ended as soon as possible (closer to the top dead center). However, in practice, when the fuel injection amount is relatively large, the above conditions cannot be completely satisfied because the fuel injection time becomes long. In addition, at the time of high engine speed, even if the fuel injection amount is small, the time of the intake stroke becomes shorter than the fuel injection time, and the same problem occurs. In addition, the piston speed also has a great influence on the formation of a uniform mixture. That is, the optimal intake stroke fuel injection start and end timings are determined by the engine load (that is, the fuel injection amount (fuel injection time)), the rotation speed, the piston speed at the time of injection, and the like.

In the compression stroke fuel injection for stratification, it is necessary to form a rich (rich) mixture layer around the spark plug without diffusing the injected fuel into the cylinder. The air-fuel ratio of the rich mixture layer formed by the injection is in a range where ignition is easy (for example, about 13 to 14 in air-fuel ratio).
Must be In order to stratify the fuel injected in the compression stroke fuel injection as a rich mixture, it is desirable to delay the injection start time as much as possible so that the injected fuel does not diffuse before ignition. However, since the compression stroke fuel injection is performed during the rise of the piston in the latter half of the compression stroke, the later the end timing of the compression stroke fuel injection, the more the injected fuel is compressed by the rise of the piston before diffusing, resulting in stratification. The air-fuel ratio of the air-fuel mixture is reduced (enriched). For this reason, when the fuel injection timing is delayed, the stratified air-fuel mixture becomes excessively rich, which may result in poor ignition and combustion.

Therefore, even in the compression stroke fuel injection, the optimal start and end timings of the fuel injection vary depending on the engine load, the rotational speed, the piston speed at the time of the injection, and the like. As described above, it is difficult for an actual engine to obtain an ideal injection timing because many factors influence the optimum fuel injection timing for both the intake stroke fuel injection and the compression stroke fuel injection. For this reason, in practice, the fuel injection timing is set such that the state of formation of the air-fuel mixture is as ideal as possible under various combinations of operating conditions (engine speed, load) by actually operating the engine while changing the speed and load. It is determined by a so-called adaptation operation that is set so as to approach the state.

Therefore, when the fuel injection amount is corrected in consideration of the fuel supplied to the cylinder by purging the fuel vapor, only the fuel injection amount is corrected while the fuel injection start or end timing is fixed. In this case, the mixture formation state set by the above-mentioned adaptation may be broken, and good combustion may not be obtained. In the present embodiment, in order to solve the above problem, when correcting the fuel injection amount according to the fuel vapor amount, both the injection start timing and the end timing are changed.

FIG. 2 is a diagram for explaining the difference between the fuel injection timing correction of the present embodiment and the conventional fuel injection timing correction.
FIG. 2 shows the fuel injection timing, and the horizontal axis represents the crank angle (CA). The line I in FIG. 2 shows the fuel injection timing (base fuel injection timing) when purging is not performed. That is, at the base fuel injection timing, the fuel injection is started at the crank angle CA1, and the crank angle C
The process ends at A2.

Lines II and III in FIG. 2 show the conventional fuel injection correction. Line II shows the case where the injection start timing is fixed, and line III shows the case where the injection end timing is fixed. That is, when performing correction to reduce the fuel injection amount (injection time) by VP according to the fuel vapor amount, the injection end timing is advanced by VP in the fixed injection start timing (II), and the injection end timing is fixed by the VP in the fixed injection end timing (III). A correction for delaying the injection start timing by VP is performed.

A line IV in FIG. 2 indicates the fuel injection correction of the present embodiment. As shown in FIG. 2 and line IV, in this embodiment, when the injection time is reduced by VP, the time VP is set to V
The operation is divided into P1 and VP2 (VP = VP1 + VP2), and an operation of delaying the injection start timing by VP1 and advancing the injection end timing by VP2 is performed. The ratio between VP1 and VP2 is set in accordance with each rotational speed and load based on an experiment using an actual engine such that the state of formation of the air-fuel mixture approaches the ideal state as much as possible. Thus, even when the fuel vapor amount is corrected, the fuel injection start timing and the fuel injection start timing are more appropriately adjusted, and good combustion can be obtained.

The details of the fuel injection control during the execution of the purge in the homogeneous mixture combustion and the stratified combustion will be described below. (1) Uniform air-fuel mixture combustion 1) First embodiment In this embodiment, an example in which fuel injection control is performed with fuel injection start timing as a reference timing will be described. In this case, the ECU 30 controls the fuel injection amount and the fuel injection timing by setting the injection start timing and the injection time of each cylinder according to the engine operating state. In this case, conventionally, when the fuel injection amount is reduced in order to correct the fuel vapor amount at the time of performing the purge, only the fuel injection time is shortened without changing the fuel injection start timing (while being fixed). The fuel injection amount was reduced. On the other hand, the present embodiment is different from the related art in that when the fuel injection time is changed, the fuel injection start timing serving as the reference timing is also corrected (delayed). As a result, both the fuel injection start timing and the fuel injection end timing are adjusted according to the fuel vapor amount, and an optimum mixture formation state can be obtained.

FIG. 3 is a flow chart for explaining the purge fuel injection control operation of this embodiment. This operation is E
This is performed as a routine executed by the CU 30 at every constant crank rotation angle. When the operation of FIG. 3 is started, in step 301, it is determined whether or not the engine is currently operating in the homogeneous mixture combustion mode (that is, any of the above-described operation modes (c) to (e)). Only when the engine is operated in the combustion mode, the fuel injection control of steps 303 to 329 is performed. If it is determined in step 301 that the engine is not currently operating in the homogeneous mixture combustion operation mode (that is, the engine is operating in the stratified combustion operation mode or the weak stratified combustion operation mode), the ECU is separately provided.
By 30, fuel injection control according to each operation mode described later is performed.

In step 303, the crank angle sensor 3
The engine speed NE and the load parameter KL calculated based on the three outputs are read. Here, as the load parameter KL, a lean air-fuel ratio uniform mixture combustion (mode
In the case of (c)), the accelerator opening ACCP detected by the accelerator opening sensor 37 is detected by the intake pressure sensor 35 in the stoichiometric air-fuel ratio or rich air-fuel ratio uniform mixture combustion (modes (d) and (e)). The detected intake pipe pressure is used.

Next, at step 305, the sensor output ratio α is calculated based on the oxygen concentration detected by the intake oxygen concentration sensor 31, and further, based on α and the current operating air-fuel ratio of the engine, the sensor corrected output ratio is calculated. A is calculated. The sensor output ratio α is the output of the intake oxygen concentration sensor 31 when the purge is not being executed, that is, the output of the intake oxygen concentration R0 when the purge is not being executed and the output of the present (purge is being executed) oxygen sensor 31 (the current output). (Intake oxygen concentration) RP, which is given as α = RP / R0.

When fuel vapor is present in the intake air, the oxygen in the intake reacts with the fuel vapor on the sensor 31 and is consumed.
Therefore, on the sensor 31, the oxygen concentration is reduced by the amount consumed for the reaction with the fuel vapor, and the sensor output becomes RP. That is, of the oxygen in the intake air, an amount of oxygen corresponding to R0 × (1−α) is consumed by the reaction with the fuel vapor. However, since the reaction between fuel and oxygen on the sensor 31 is an equivalent reaction, it corresponds to combustion of an air-fuel mixture with an excess air ratio λ = 1. For this reason, for example, the lean air-fuel ratio (λ> 1)
When the fuel vapor is burned, (1
It is necessary to allocate oxygen in an amount of -α) × λ. That is, in order to maintain the excess air ratio λ of the combustion chamber in the engine, the amount of oxygen that can be allocated to the fuel supplied by the fuel injection decreases to R0 × (1− (1−α) × λ). . In other words, if λ in the combustion chamber is kept the same in the presence of the fuel vapor, the amount of oxygen that can be used for combustion of the fuel injected by the fuel injection becomes the oxygen concentration (R0) when the fuel vapor is not present in the intake air. (1- (1-α) ×
λ) times.

In this case, the amount of oxygen available for burning the fuel injected by the fuel injection is (1- (1-α)
× λ) times, so that even if fuel paper is present in the intake air, in order to maintain the same combustion air-fuel ratio, the fuel injection amount also needs to be reduced by the same amount as the oxygen reduction rate. Therefore, when the output ratio of the intake oxygen concentration sensor during the purge is α, the fuel injection amount is set to (1− (1−α) ×
If the amount is reduced by λ), the same air-fuel ratio as in the case where there is no vapor can be maintained.

In the present embodiment, it is calculated using α and λ (λ = operating air-fuel ratio / stoichiometric air-fuel ratio) (1− (1−α)).
× λ) is defined as the sensor corrected output ratio A (A = (1
− (1-α) × λ)). In order to maintain the same air-fuel ratio at the time of performing the purge, the fuel injection amount at the time of performing the purge may be corrected to a value obtained by multiplying the base fuel injection amount by the sensor correction output ratio A.

However, in the present embodiment, when the purge is performed, the fuel is shifted to the lean side as compared with the case where the purge is not performed, so that the fuel injection amount is corrected to the base fuel injection amount as described below. The value is corrected to a value smaller than the value obtained by multiplying the ratio A. That is, after calculating the corrected output ratio A of the oxygen concentration sensor in step 305, in step 307, the value of vapor concentration B (hereinafter, referred to as “vapor equivalent value B”) during intake is expressed as B = 1−A = (1 The vapor equivalent value B is a ratio of the amount of fuel vapor during fuel intake to the base fuel injection amount, and is a value corresponding to the vapor concentration.

After calculating the vapor equivalent value B in step 307, in step 309, the fuel injection amount reduction constant γ is calculated based on the value of the vapor equivalent value B. As described above, γ is set to a smaller value as the value of B is larger.
In the present embodiment, the relationship between the vapor equivalent value B at the time of homogeneous mixture combustion and the optimal value of the weight loss constant γ is previously obtained by experiments, and stored in the ROM of the ECU 30 in the form of a numerical map using B. Have been. In step 309, γ is determined from the value of the vapor equivalent value B using this numerical map.

Further, in step 311, the value of the fuel injection amount reduction coefficient D is calculated by using γ and B obtained above.
D = γ−B. As will be described later (step 327), in the present embodiment, the actual fuel injection amount is corrected to a value obtained by multiplying the base fuel injection amount TAU by the reduction coefficient D. In step 313, the value of the adjustment coefficient E used for correcting the fuel injection timing is calculated as E = 1-D. As described above, since D is a reduction coefficient of the fuel injection amount, the value of the adjustment coefficient E is a fuel injection time (ratio) to be reduced in order to reduce the fuel injection amount, that is, VP1 in FIG.
And VP2 and the ratio of the fuel injection time before the decrease.

Next, in steps 315 and 317, the base fuel injection start timing I is determined by using the engine speed NE and the load parameter KL read in step 303 from the numerical table stored in the ROM of the ECU 30 in advance.
NJT and base fuel injection amount TAU are calculated respectively. Steps 319 to 325 show an operation for calculating the fuel injection timing correction amount β. In the present embodiment, since the fuel injection control is performed based on the fuel injection start timing,
The correction amount β represents a retard amount (crank angle) at the start of fuel injection.

The injection timing correction (retardation) amount β is shown in FIG.
This is a crank angle corresponding to VP1 of V, and is determined according to the adjustment coefficient E, the engine speed NE, and the engine load KL.
That is, in step 319, the vapor correction amount KINJ
VP is determined based on the adjustment coefficient E. The vapor correction amount KINJVP corresponds to the fuel injection time to be shortened due to the presence of the fuel vapor, and corresponds to the fuel injection shortening time VP (= VP1 + VP2) of line IV in FIG. FIG.
9 is a graph showing a relationship between INJVP and an adjustment coefficient E.
KINJVP (fuel injection shortening time) increases almost in proportion to the adjustment coefficient E.

In steps 321 and 323, the rotation speed correction coefficient KNE1 and the load correction coefficient KKL1 are determined based on the engine speed NE and the load parameter KL, respectively. The correction coefficients KNE1 and KKL1 correspond to the fuel injection reduction time KINJVP by the fuel vapor.
Is a coefficient for determining how much is allocated to the fuel injection start delay side (that is, VP1 of line IV in FIG. 2). In practice, the relationship between FIGS. 5 and 6 is determined based on experiments. 5 and 6 show KNE1 and NE, respectively.
It is a graph which shows an example of the relation between KKL1 and KL. KN
E1 and KKL1 increase as NE and KL increase, respectively.

As described above, in the fuel injection during the intake stroke, the fuel injection start timing is determined as the engine speed increases and the load increases.
The larger the (fuel injection amount) is, the more the fuel injection time needs to be set to the advanced side to secure the fuel injection time. However, this fuel injection start timing may not always be optimal from the viewpoint of forming a uniform mixture. For this reason, in the present embodiment, when the engine speed or the load is high, the injection start timing is further retarded so that the fuel injection start timing approaches the optimum timing.

The injection start delay amount β (β ≧ 0) is calculated in step 325 as β = KINJVP × KNE1 × KKL1. After calculating the retardation amount β, in step 327, the base fuel injection amount (fuel injection time) TAU is reduced and corrected based on the reduction coefficient E, and the actual fuel injection amount TAU1 is calculated as
It is calculated as TAU1 = TAU × D.

In step 329, the actual fuel injection start timing INJT1 is changed to the base fuel injection start timing I
It is calculated as INJT1 = INJT + β using NJT. Thus, in a separately executed fuel injection operation, the fuel injection from the fuel injection valve is started when the crank reaches INJT1, and the fuel injection is stopped when the fuel injection continues for TAU1 (millisecond). Is controlled.

In this embodiment, the fuel injection start timing INJ
Since T is represented by the crank angle after intake top dead center (ATDC), the injection start timing is delayed by the crank angle β according to the adjustment coefficient E as described above. Also, the fuel injection time TAU
Therefore, the fuel injection end timing is also advanced according to the adjustment coefficient E, and both the fuel injection start timing and the fuel injection end timing are set to appropriate values according to the fuel injection amount. Will be set.

2) Second Embodiment In the first embodiment, the injection start delay amount β is directly calculated from the adjustment coefficient E by using the correction coefficients KNE1 and KKL1, but in the present embodiment, The difference from the first embodiment is that a fuel injection start delay time (millisecond) is calculated from the adjustment coefficient E, and this delay time is converted into a delay amount β (crank angle). Thus, in the present embodiment, it is possible to set the fuel injection start timing more accurately.

FIG. 7 is a flowchart for explaining the fuel injection control operation of this embodiment. This operation is performed by the ECU 30
Is performed as a routine executed at every constant crank rotation angle. In steps 701 to 715 in FIG. 7, the sensor correction output ratio A, the decrease count D: the adjustment coefficient E, and the base fuel injection start timing INJT are calculated in the same manner as in the operation in FIG. 3. Steps 701 to 715 correspond to step 301 in FIG.
Since the operation is the same as that of steps 315 to 315, the description is omitted.

After executing steps 701 to 715, in this embodiment, at step 717, the base fuel injection time TAU (millisecond) is calculated based on the rotational speed NE and the load parameter KL. In step 719, the amount TAUB for which the fuel injection time is to be reduced is calculated as TAUB = TAU ×
It is calculated as E. TAUB is a time (millisecond) corresponding to the time VP in FIG.

Then, in step 721, the shortened time TA
The correction coefficient KINJT1 representing the ratio of the UB to be allocated to the fuel injection start delay is determined by the engine speed NE and the load K
It is calculated based on L. In the present embodiment, the optimum value of the correction coefficient KINJT1 is set in advance based on experiments, and is stored in the ROM of the ECU 30 in the form of a two-dimensional numerical table using NE and KL. Step 72
In step 1, based on the values of NE and KL read in step 703, the correction coefficient KINJT1
Is determined.

In step 723, the conversion coefficient C for converting the injection start delay time (millisecond) into the crank rotation angle is calculated as C = (60 × 1000) based on the current engine speed NE (RPM). ) / (NE × 360). Then, in step 725, the fuel injection start delay time is calculated as TAUB × KINJT1, and further converted into a crank rotation angle, that is, an injection start delay amount (CA) with respect to the base fuel injection start timing, using the above conversion coefficient. You. In step 727, the actual fuel injection time TAU1 is calculated as TAU1 = TAU × D, and the actual fuel injection start timing INJT1 is calculated.
Are set as INJT1 = INJT + β. Thus, similarly to the first embodiment, both the fuel injection start timing and the end timing are set to more appropriate values based on the fuel vapor amount.

In the first and second embodiments, the case where the fuel injection control is performed based on the fuel injection start timing is described. However, the same control is performed when the fuel injection control is performed based on the fuel injection end timing. It is also possible. In this case, of the shortened fuel injection time due to the presence of the fuel vapor, a correction coefficient representing the amount to be distributed to the advance angle of the fuel injection end timing is stored in the ROM of the ECU 30 in advance,
The advance amount of the fuel injection end timing may be calculated by the same operation as in FIGS.

(2) Stratified Combustion Next, a description will be given of the correction by the purge of the compression stroke injection during the stratified combustion operation. In the compression stroke fuel injection during stratified charge combustion, the fuel injection timing is set so that fuel is injected into the cylinder-free air during the compression stroke to form a rich mixture layer. On the other hand, if purging is performed during stratified charge combustion, in the compression stroke injection, fuel is injected into a uniform mixture including fuel vapor in the cylinder. For this reason, in the purge during stratified combustion, even if the fuel injection amount is reduced, if the fuel injection is performed at the same timing as usual, the air-fuel ratio of the formed air-fuel mixture layer becomes rich, and ignition and combustion deteriorate. May be. Therefore, in the present embodiment, when purging is performed during stratified charge combustion, the end timing of the compression stroke fuel injection is advanced to make the injected fuel more easily diffused, thereby preventing the mixture from being enriched.

1) Third Embodiment FIG. 8 is a flowchart illustrating a fuel injection control operation during stratified charge combustion. This operation is executed by the ECU 30 at every constant crank rotation angle. In the present embodiment, the case where the fuel injection control is performed based on the fuel injection end timing is shown. Further, the calculation of the fuel injection amount and the advance amount of the fuel injection end timing is described in the above-mentioned second case in the case of the homogeneous mixture combustion.
This is based on almost the same concept as the embodiment.

That is, in step 801 in FIG. 8, it is determined whether or not the engine operation in the stratified combustion mode is currently performed. Only when the operation in the stratified combustion mode is currently performed, the compression strokes in steps 803 to 829 are performed. Fuel injection control is performed. If it is determined in step 801 that the operation in the stratified combustion mode is not currently being performed (that is, if the operation in the homogeneous mixture combustion mode or the weak stratified combustion mode is being performed), the fuel corresponding to each mode is separately set. Injection control is performed.

In step 803, the engine speed NE and the load parameters described above (for the stratified combustion operation mode,
The accelerator opening ACCP) KL is read, and step 80 is executed.
In steps 5 and 807, the sensor corrected output ratio A and the vapor equivalent value B are calculated from the output of the intake oxygen concentration sensor 31. These operations are the same as those in the first and second embodiments.

After calculating the vapor equivalent value B in step 807, in step 809, a reduction constant γ of the fuel injection amount is calculated based on the value of the vapor equivalent value B. The value of the reduction constant γ calculated here is different from Step 309 in FIG. 3 in that it is an optimum value of γ during stratified combustion obtained in advance by an experiment. Steps 811 and 813
Then, the values of the reduction coefficient D and the adjustment coefficient E of the fuel injection amount are calculated in the same manner as in steps 311 and 313 in FIG.

Next, in steps 815 and 817, the base fuel injection end timing INJTF and the fuel injection time TA during the stratified combustion are obtained from the numerical table stored in the ROM of the ECU 30 in advance by using the rotational speed NE and the load parameter KL.
U (millisecond) is determined. In step 819, the injection time reduction amount TAUB (millisecond) for fuel vapor correction is calculated as TAUB = TAU × E.

Then, at step 821, a fuel injection timing correction coefficient KINJT2 is calculated. KINJT2
Is a correction coefficient having the same role as KINJT1 in the above-described second embodiment. In this embodiment, of the fuel injection time reduction (corresponding to VP in FIG. 2), the early end side of the fuel injection end timing (Corresponding to VP2 in FIG. 2). KINJT2 is set for each combination of each engine speed NE and load KL based on an experiment, and is stored in the ROM of the ECU 30 in the form of a two-dimensional numerical table using NE and KL.

After calculating the correction coefficient KINJT2 in step 821, a conversion coefficient C for converting the injection time into the crank rotation angle is obtained in step 823, and in step 825, the advance angle β of the fuel injection end timing is calculated. Steps 823 and 825 are the same operations as steps 723 and 725 in FIG. Further, in step 827, the actual fuel injection amount TAU2 of the compression stroke fuel injection is calculated using the sensor corrected output ratio A and the compression stroke base fuel injection amount TAU.
It is calculated as TAU2 = TAU × D. In step 829, the actual fuel injection end timing IN is calculated using the fuel injection end timing advance amount β calculated in step 825 and the base injection end timing INJTF of the compression stroke fuel injection.
JTF2 is calculated as INJTF2 = INJTF + β. The compression stroke fuel injection end timing INJTF
Is defined by the crank angle (BTDC) up to the compression top dead center, so that INJTF2 is the base injection end timing INJT
Becomes earlier than F by β.

Thus, the compression stroke fuel injection starts before TAU2 (millisecond) when the crank angle reaches INJTF2, and ends when the crank angle reaches INJTF2. Is set to a more appropriate value.

(3) Weakly Stratified Combustion Next, the fuel injection control during the execution of the purge in the weakly stratified combustion (the above-described operation mode (b)) will be described.

In the operation in the weak stratified charge combustion mode, both the intake stroke fuel injection for forming a homogeneous mixture and the compression stroke fuel injection for stratifying the mixture are performed. The control of the fuel injection timing and the fuel injection timing is basically the fuel injection control (combustion stroke fuel injection) during the homogeneous mixture combustion described above and the fuel injection control (compression stroke fuel injection) during stratified combustion.
And will do both. As for the fuel injection amount, the fuel equivalent to the fuel supplied as the fuel vapor is reduced from both the intake stroke fuel injection and the compression stroke fuel injection, and the total fuel amount including the corrected fuel vapor amount is calculated. Is smaller than the fuel injection amount before correction.

However, while the intake stroke fuel injection and the compression stroke fuel injection are conventionally reduced at the same ratio according to the fuel vapor amount, in the present embodiment, the intake stroke fuel injection amount and the compression stroke fuel injection are reduced. The difference from the conventional fuel injection amount correction is that the injection amount is reduced at a different ratio according to the fuel vapor amount. That is, the fuel supplied to the cylinder as the fuel vapor diffuses uniformly into the intake air to form a uniform mixture. For this reason, if the compression stroke fuel injection amount is reduced to correct the fuel amount supplied as the fuel vapor, the fuel, which would have been stratified by the compression stroke fuel injection, is supplied to the cylinder as a uniform mixture. As a result, the stratified mixture formed by the fuel injection in the compression stroke may become lean.

Therefore, in this embodiment, when correcting the fuel vapor amount, the intake stroke fuel injection amount is reduced as much as possible so that the compression stroke fuel injection amount is not reduced. Only when a large amount of fuel vapor exceeding the amount is supplied and the air-fuel ratio cannot be corrected to a leaner value than when purging is not performed even if the intake stroke fuel injection amount is reduced to zero, the compression stroke fuel injection amount is reduced. I try to lose weight. As a result, the air-fuel ratio of the stratified air-fuel mixture formed by the compression stroke fuel injection is maintained at an optimum value for ignition and combustion regardless of the presence or absence of the purge, and deterioration of combustion is prevented.

4) Fourth Embodiment FIGS. 9 and 10 are flow charts for explaining the fuel injection control operation at the time of executing the purge during the weak stratification mode operation. This operation is executed by the ECU 30 at every constant crank rotation angle.

In the present embodiment, the intake oxygen concentration sensor 31
A reduction coefficient D is obtained from a vapor-equivalent value B in the intake air calculated from the corrected output ratio A. When the fuel amount corresponding to D is smaller than the intake stroke fuel injection amount, the entire amount of fuel corresponding to the reduction coefficient D is taken in. It is subtracted from the stroke fuel injection amount, and the compression stroke fuel injection amount is not reduced. When the fuel amount corresponding to the reduction coefficient D becomes equal to or more than the intake stroke fuel injection amount, the intake stroke fuel injection is stopped, and the remaining excess fuel amount is adjusted by reducing the compression stroke fuel injection amount. I am trying to do it.

When the operation shown in FIGS. 9 and 10 starts, it is determined in step 901 whether or not the engine is currently operating in the weak stratified combustion mode. Only the operations of step 903 and subsequent steps are performed. When the engine is currently operating in an operation mode other than the weak stratified charge combustion mode (homogeneous mixture combustion mode or stratified charge combustion mode), the fuel injection control for uniform mixture combustion or stratified charge combustion described above is executed separately. Is done.

Next, at step 903, the engine speed N
E and load parameter KL (in this case, accelerator opening A
CCP) is read. In steps 904 and 905, the base total fuel injection amount (the sum of the intake stroke fuel injection amount and the compression stroke fuel injection amount when the purge is not executed) TAU and the intake stroke injection rate F are determined. Here, the intake stroke injection rate F (F ≦ 1) is a ratio of the intake stroke fuel injection amount to the total fuel injection amount TAU when the purge is not performed (intake stroke fuel injection amount / (intake stroke fuel injection amount). + Compression stroke fuel injection amount)).

In this embodiment, the engine is operated in advance in the weak stratified charge combustion mode, and the optimal total fuel injection amount TAU and intake stroke injection rate F in each combination of the engine speed NE and the load KL are experimentally determined. And stored in advance in the ROM of the ECU 30 in the form of a two-dimensional numerical table using NE and KL. In steps 904 and 905, the base total fuel injection amount TAU and the intake stroke injection rate F are determined from the respective numerical tables using the current rotational speed NE and the load KL read in step 903.

In steps 907 to 915, the sensor correction output ratio A is calculated from the output of the intake oxygen concentration sensor 31, and the vapor equivalent value B, the reduction coefficient D, and the adjustment coefficient E are calculated as B = 1-A, D, respectively. = Γ-B, E =
1-D. Next, at step 917, it is determined whether or not the fuel amount corresponding to the current adjustment coefficient E is equal to or less than the base intake stroke fuel injection amount. As described above, since the adjustment coefficient E indicates the ratio of the fuel injection time to be reduced as a whole, the fuel amount to be reduced as a whole is represented by TAU × E. On the other hand, since the base intake stroke fuel injection amount is represented by TAU × F, by comparing the adjustment coefficient E with the base intake stroke injection rate F, the total amount of fuel to be reduced this time is calculated from the intake stroke fuel injection amount. It can be determined whether reduction can be achieved.

If E ≦ F in step 917, the total amount of fuel to be reduced from the intake stroke fuel injection amount can be reduced. Therefore, in this case, step 91
An intake stroke fuel injection correction operation from 9 to 931 is performed.
The intake stroke fuel injection correction operation in steps 919 to 931 is basically the same as the fuel injection correction in the homogeneous mixture combustion mode in FIG.

That is, in step 919, the base intake stroke fuel injection end timing INJT1F is determined based on a numerical table stored in advance in the ROM of the ECU 30 using the engine speed NE and the load parameter KL. The reduction amount TAUB1 of the intake stroke fuel injection time required to reduce the total amount of fuel to be used from the intake stroke fuel injection amount is the total fuel injection amount (time) TAU.
Is calculated as TAUB1 = TAU × E using the adjustment coefficient E. In step 923, the engine speed N
Based on E and the load KL, the injection timing correction coefficient KIN is obtained from a numerical table stored in the ROM of the ECU 30 in advance.
JTF is required. The correction coefficient KINJT1F is a coefficient that determines how much of the shortened intake stroke fuel injection time is to be distributed to the side that advances the injection end timing, and is based on experiments using an actual engine to determine each engine speed NE and load KL.
The optimal value is determined for the combination with
EC in advance in the form of a two-dimensional numerical table using E and KL
It is stored in the ROM of U30.

Next, in step 925, a conversion coefficient C for converting the injection time into the crank rotation angle is calculated based on the current engine speed NE. In step 927,
When the advance amount β1 (CA) at the end of the intake stroke fuel injection is T
Using AUB1 and correction coefficients KINJT1F and C, β
1 = (TAUB1 × KINJT1F) / C

In step 927, the actual intake stroke fuel injection amount (time) TAU1 for correcting the fuel vapor during intake is calculated as TAU1 = TAU × (FE). In step 931, the intake stroke fuel injection end timing INJTF1 (CA) is changed to INJTF1
= INJT1F-β1. Since the intake step fuel injection end timing INJTF1 is represented by the crank rotation angle (ATDC) after the intake top dead center, the fuel injection end timing is advanced by β1 in the crank angle (CA). Become.

Thus, the start and end timings of the fuel injection in the intake stroke are adjusted to appropriate values according to the fuel vapor amount. When the entire fuel amount to be reduced can be corrected by reducing the intake stroke fuel injection amount, the air-fuel ratio of the formed homogeneous mixture becomes leaner than when no purge is performed, and the compression stroke fuel No correction is made for the injection amount and injection timing of the injection. That is, in this case, the compression stroke fuel injection amount and the injection timing are set to the base injection amount and the base injection timing, respectively.

Next, correction in the case where E> F in step 917, that is, the case where the fuel amount to be reduced is larger than the intake stroke fuel injection amount, will be described. As described above, in this case, the fuel injection in the intake stroke is stopped (that is, TAU1 = 0), and the fuel vapor amount (EF) that cannot be corrected even more is corrected by reducing the fuel injection amount in the compression stroke.

That is, if E> F in step 917, the process proceeds to step 933 in FIG. 10, where the intake stroke fuel injection amount TAU1 is set to TAU1 = 0, and the intake stroke fuel injection is stopped. In step 935, the ECU 30 determines the value of the engine speed NE and the load parameter KL.
The base fuel injection end timing INJT2F of the compression stroke fuel injection is determined based on the numerical table stored in the ROM of FIG.

Then, in steps 937 to 947,
The compression stroke fuel injection amount is reduced by (EF), and the compression stroke fuel injection end timing INJTF2 is advanced by an amount β2 according to the fuel injection reduction amount (EF), the engine speed NE, and the load KL. Set the angle. Steps 937 to 947 are the same operations as steps 819 to 829 in the fuel injection correction at the time of stratified combustion in FIG. That is, (E−F) in step 937 is a reduction amount of the compression stroke fuel injection amount, and corresponds to E in step 819 in FIG. 8, and TAU2 = TAU in step 945.
× ((1−F) − (E−F)) represents step 827 in FIG.
Corresponds to TAU2 = TAU × D.

Accordingly, even when the fuel reduction amount for the fuel vapor correction is larger than the intake stroke fuel injection amount and the compression stroke fuel injection needs to be corrected, the injection start timing and the end of the compression stroke fuel injection are required. The timing is set to an optimal value according to the fuel vapor amount.

5) Fifth Embodiment Next, another embodiment of the fuel injection correction operation at the time of executing the purge in the weak stratified charge combustion operation will be described. The fourth mentioned above
In the embodiment, the intake stroke fuel injection is stopped when the fuel amount to be reduced during the intake becomes equal to or more than the intake stroke fuel injection amount, but in the present embodiment, the fuel amount to be reduced is the intake stroke fuel injection amount. Even if the amount exceeds the predetermined amount, fuel injection of a predetermined amount is performed in the intake stroke, and the total of the fuel amount injected in the intake stroke and the fuel amount to be originally reduced is reduced from the compressed fuel injection amount to reduce the overall engine operating air-fuel ratio. Lean compared to not performing the purge. In this case, the predetermined amount of intake fuel injection is performed in the latter half of the intake stroke, so that the injected fuel is not uniformly diffused in the cylinder.

As a result, the fuel injected during the intake stroke forms a mixture layer having a slightly lower air-fuel ratio (richer) than the uniform mixture in the homogeneous mixture including the fuel vapor in the cylinder. In the compression stroke fuel injection, fuel is injected into the air-fuel mixture having a slightly lower air-fuel ratio to form a rich air-fuel mixture. Thereby, an air-fuel mixture layer having an intermediate air-fuel ratio of the fuel injected in the intake stroke is interposed between a rich air-fuel mixture layer formed by the compression stroke fuel injection and a (lean) uniform air-fuel ratio having a high air-fuel ratio. I will be. For this reason, when ignition occurs in the rich air-fuel mixture layer, the flame smoothly propagates through the intermediate air-fuel ratio layer to the lean homogeneous air-fuel mixture. Normal,
In stratified charge combustion, the difference in air-fuel ratio between the stratified rich air-fuel ratio air-fuel mixture and the homogeneous air-fuel mixture in the cylinder increases, and it may be difficult for the flame to smoothly propagate from the rich air-fuel ratio layer to the uniform air-fuel mixture. However, in the present embodiment, as described above, the air-fuel mixture layer having the intermediate air-fuel ratio is provided between the rich air-fuel ratio layer and the uniform air-fuel mixture layer, so that the flame propagates smoothly and the weak stratified combustion mode is used. The combustion state can be improved.

FIG. 11 is a flowchart specifically illustrating the fuel injection control according to the present embodiment. The operation in FIG. 11 is executed together with the operation in FIG. 9 instead of the operation in FIG. 10 of the fourth embodiment. That is, in the present embodiment, when the fuel amount to be reduced in step 917 in FIG. 9 is larger than the intake stroke base fuel injection amount (E> F),
In FIG. 11, the routine proceeds to step 1101, and the intake stroke fuel injection amount T
AU1 is calculated as TAU1 = TAU × G. Here, TAU is the base total fuel injection amount obtained in step 905 in FIG. In the present embodiment, G is a relatively small constant value (for example, a value of about 0.05 to 0.1).

Further, at step 1103, the intake stroke fuel injection end timing INJTF1 is determined based on the engine speed NE and the load KL. E> F at step 911
Is different from the normal intake stroke fuel injection for forming a homogeneous mixture, the intermediate fuel having an air-fuel ratio lower than that of the surrounding mixture by preventing the injected fuel from diffusing uniformly. It is necessary to form a fuel-fuel mixture layer. Therefore, the injection end timing I set in step 1103
NJTF1 is the INJ set in step 931 in FIG.
Unlike TF1, it is set in the latter half of the intake stroke. In the present embodiment, the optimal fuel injection end timing for forming the intermediate mixture layer is set in advance according to each combination of the engine speed NE and the engine load KL based on an experiment.
It is stored in advance in the ROM of the ECU 30 in the form of a two-dimensional numerical table using E and the load KL as parameters. In step 1103, the end timing INJTF1 of the intake stroke fuel injection is set based on NE and KL read in step 903 in FIG.

Steps 1105 to 1117
Shows an operation for setting the injection amount of the compression stroke fuel injection and the injection end timing. The operations in steps 1105 to 1117 are as follows:
Except for the fact that the fuel injection amount is further reduced by G in steps 1107 and 1115, the operation is the same as that in steps 935 to 947 in FIG.

As described above, even when the amount of fuel to be reduced according to the amount of fuel vapor during intake is larger than the fuel injection amount in the intake stroke, a small amount of fuel is injected in the latter half of the intake stroke, so that no purge is performed. As compared with the case, it is possible to further improve the combustion state at the time of performing the purge while maintaining the air-fuel ratio lean.

[0122]

According to the present invention, when purging is performed in the in-cylinder fuel injection type spark ignition engine, combustion is performed by effectively utilizing the fuel vapor as a uniform mixture supplied by the purge. By improving the state, it is possible to shift the engine air-fuel ratio at the time of performing the purge to the lean side as compared with the case where the purge is not performed, so that both the fuel consumption and the exhaust property are maintained while maintaining the same output. Has a common effect that can be improved.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of an embodiment when the present invention is applied to an internal combustion engine for a vehicle.

FIG. 2 is a timing chart illustrating correction of fuel injection timing in the embodiment of FIG. 1;

FIG. 3 is a flowchart illustrating a first embodiment of a fuel injection control operation of the present invention.

FIG. 4 is a graph showing setting of coefficients used for the control operation of FIG. 3;

FIG. 5 is a graph showing setting of coefficients used for the control operation of FIG. 3;

FIG. 6 is a graph showing setting of coefficients used for the control operation of FIG. 3;

FIG. 7 is a flowchart illustrating a second embodiment of the fuel injection control operation of the present invention.

FIG. 8 is a flowchart illustrating a third embodiment of the fuel injection control operation of the present invention.

FIG. 9 is a part of a flowchart illustrating a fourth embodiment of the fuel injection control operation of the present invention.

FIG. 10 is a part of a flowchart illustrating a fourth embodiment of the fuel injection control operation of the present invention.

FIG. 11 is a part of a flowchart illustrating a fifth embodiment of the fuel injection control operation of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Internal combustion engine 10 ... Intake passage 30 ... Electronic control unit (ECU) 31 ... Intake oxygen concentration sensor 40 ... Evaporated fuel purge device 111-114 ... In-cylinder fuel injection valve

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) F02D 43/00 F02D 43/00 301J 301M F term (reference) 3G084 AA03 BA05 BA13 BA15 BA17 BA20 BA27 CA03 CA04 DA10 EA04 EA11 EB08 EB11 EC01 EC03 FA00 FA10 FA11 FA33 FA38 3G301 HA01 HA04 HA06 HA13 HA14 HA16 HA17 JA21 KA06 KA23 LA03 LA05 LB04 LC04 MA11 MA19 NA06 NA08 NB14 NC02 ND01 NE06 NE13 NE14 NE15 NE19 PA00A PA00Z PA07A PA07Z PD04

Claims (7)

[Claims] An evaporative fuel purge device for supplying evaporative fuel in a fuel tank to an engine intake passage; a vapor detector for detecting the evaporative fuel concentration in engine intake air; and a cylinder for directly injecting fuel into a cylinder. A fuel injection control device for an internal combustion engine, comprising: an internal fuel injection valve; and a fuel injection setting unit configured to set a fuel injection amount from the in-cylinder fuel injection valve based on an engine operating state. When evaporative fuel is supplied from the fuel purge device to the engine intake passage, the amount of evaporative fuel supplied to the engine combustion chamber and fuel injection into the engine combustion chamber based on the evaporative fuel concentration detected by the vapor detection means. Fuel injection amount correction for correcting the fuel injection amount set by the fuel injection setting means so that the total of the supplied fuel amount and the fuel injection amount is smaller than the fuel injection amount set by the setting means. The fuel injection control apparatus for an internal combustion engine having a stage, a. 2. The fuel injection amount correction means according to claim 1, wherein the difference between the fuel injection amount set by the setting means and the total fuel amount increases as the detected fuel vapor concentration increases. The fuel injection control device according to claim 1, wherein the set fuel injection amount is corrected. 3. The fuel injection control device according to claim 1, wherein the fuel injection valve performs a fuel injection in an intake stroke of each cylinder to perform a uniform combustion mode fuel injection in which a uniform mixture is formed in the cylinder. Fuel injection control device. 4. The fuel injection control device according to claim 1, wherein fuel is injected from the fuel injection valve in an intake stroke of each cylinder to form a uniform mixture in the cylinder and a compression stroke of each cylinder. A fuel injection control device for performing a weak stratified charge combustion mode fuel injection, wherein the fuel mixture is injected into the homogeneous mixture to form a layer of a mixture having a low air-fuel ratio in the uniform mixture. 5. The fuel injection control device according to claim 1, wherein fuel is injected from the fuel injection valve in a compression stroke of each cylinder to form a layer of a combustible mixture in air in the cylinder. A fuel injection control device that performs mode fuel injection. 6. The fuel injection control device according to claim 1, wherein fuel is injected from the fuel injection valve in an intake stroke of each cylinder in accordance with an engine operating condition to form a uniform mixture in the cylinder. In the uniform combustion mode fuel injection, fuel injection is performed from the fuel injection valve in the intake stroke of each cylinder to form a uniform mixture in the cylinder, and fuel injection is performed in the compression stroke of each cylinder. A low stratified charge combustion mode fuel injection to form a layer of a mixture having a low air-fuel ratio; A fuel injection control device that performs fuel injection by selecting one of the following: stratified combustion mode fuel injection. 7. A fuel injection timing setting means for setting a fuel injection timing from the in-cylinder fuel injection valve based on an engine operating state, wherein the fuel injection control means further comprises a fuel injection timing detecting means for detecting the fuel vapor concentration. 2. The fuel injection control device according to claim 1, wherein both the fuel injection start timing and the end timing set by the fuel injection timing setting means are changed accordingly.