JP7408961B2 - Diesel engine control device - Google Patents

Description

The present invention relates to a device for controlling a diesel engine that includes a cylinder, a piston that is reciprocatably housed in the cylinder, and an injector that injects fuel containing light oil into a combustion chamber that is a space above the piston.

As an example of the above-mentioned diesel engine, the one shown in Patent Document 1 below is known. In the diesel engine of Patent Document 1, injection patterns that combine pre-injection and after-injection with main injection are determined differently depending on operating conditions, and when fuel is injected by each injection pattern, the fuel pressure in the injector is controlled. The injection timing and injection period are adjusted based on the detected value of the pressure sensor.

For example, in an injection pattern that includes main injection and after injection, the pressure pulsations that occur with fuel injection are detected by a pressure sensor, and based on the detected pressure pulsations, the interval from main injection to after injection (injection interval ) is adjusted. This is said to reduce the influence of pressure pulsations in the fuel pressure on the after-injection, thereby improving the accuracy of adjusting the injection amount of the after-injection.

Japanese Patent Application Publication No. 2011-190725

Here, if the interval from main injection to after injection is too short, fuel by after injection will be superimposed on the combustion gas based on main injection, and the fuel by after injection will be in an oxygen-deficient environment. As a result of combustion, soot is likely to be generated. Therefore, in order to reliably avoid the generation of such soot, it is conceivable to make the interval from the main injection to the after injection sufficiently long. However, if the interval becomes too long, the proportion of the combustion energy based on after-injection that is used as work decreases, resulting in poor fuel efficiency.

In order to prevent both the generation of soot and the deterioration of fuel efficiency as described above, it is desirable to set the timing of after-injection as early as possible during the period when sufficient air (oxygen) is available. . However, the timing of after-injection that can meet such demands is thought to change from time to time depending on the flow state of gas in the combustion chamber and the like. Therefore, it is proposed to grasp the state quantities that affect the flow of gas in the combustion chamber, and to determine the appropriate timing of after-injection each time based on the results.

However, the diesel engine disclosed in Patent Document 1 cannot meet the above proposals. That is, in Patent Document 1, although there is room for correction based on pressure pulsations, basically the interval from main injection to after injection is determined based on an injection pattern that is predetermined for each engine operating condition. In other words, the timing of the after injection is determined according to a basic interval determined in advance experimentally. Therefore, in Patent Document 1, it is impossible to determine the optimal after-injection timing according to the constantly changing engine conditions.

The present invention has been made in view of the above circumstances, and aims to sufficiently suppress soot generation by increasing the air utilization rate of fuel injected by after-injection while maintaining relatively good fuel efficiency. The purpose of the present invention is to provide a diesel engine control device that can perform the following tasks.

In order to solve the above problems, the present invention includes a cylinder, a piston housed in the cylinder so as to be able to reciprocate, and an injector that injects fuel containing light oil into a combustion chamber that is a space above the piston. A device for controlling a diesel engine, comprising an injection control unit that controls the injector so that fuel is injected at a plurality of timings set from a compression stroke to an expansion stroke when operating in a predetermined operating range. , the piston has a downwardly recessed cavity on its crown surface, and a bottom portion formed such that the height becomes lower toward the radially outer side as a wall surface defining the cavity, and a bottom portion that is formed radially outwardly from the bottom portion. A curved outer circumferential portion is formed and concave so as to be convex radially outward in a cross-sectional view including the cylinder axis, and a curved outer circumferential portion is formed above the outer circumferential portion and is convex radially inward in a cross-sectional view including the cylinder axis. The injector has a plurality of nozzle holes in the ceiling of the combustion chamber at a position facing the center of the cavity, and the injector has a plurality of nozzle holes extending radially from each nozzle hole. The injection control unit is provided to inject fuel diagonally downward toward the outside in the radial direction, and the injection control unit is configured to inject the fuel in a direction that injects the largest proportion of the total amount of fuel injected during one combustion cycle during operation in the predetermined operating region. a main injection that injects fuel and directs the injected fuel toward the lip portion to redirect at least a portion of the fuel downward from the lip portion; and at a predetermined time during the expansion stroke that is delayed from the main injection. The injector is made to perform an after injection that injects a smaller amount of fuel than the main injection, and the injection interval time, which is the time from the end of the main injection to the start of the after injection, becomes shorter as the engine speed increases. In addition, the reduction rate of the injection interval time with respect to an increase in engine speed is increased as the engine speed is higher, except when the engine load corresponds to a maximum value within the predetermined operating range. (Claim 1)

The fuel spray injected by the main injection swirls along the lip, outer periphery, and bottom walls of the cavity to form a vertical vortex, and returns to a specific position on the injection axis of the injector. . In other words, the oxygen concentration at the specific position varies greatly depending on the swirling flow of the fuel spray caused by the main injection. Therefore, in order to increase the air utilization rate of the fuel injected by after-injection, the timing when the fuel spray by the after-injection reaches the specific position and the timing when the oxygen concentration at the specific position becomes high (hereinafter referred to as oxygen arrival It is necessary to roughly match the period (also called the period).

Through research by the inventors of the present application, it has been found that the timing of oxygen arrival, which changes due to the swirling of the fuel spray forming the above-mentioned vertical vortex, becomes earlier as the engine speed increases. As a control that takes this point into account, in the present invention, the injection interval time, which is the time from the end of main injection to the start of after injection, is adjusted so that it becomes shorter as the engine speed increases. It is possible to cause the fuel spray by after-injection to reach the specific position at a time when the oxygen concentration of the fuel spray becomes high, and the air utilization rate of the fuel spray can be increased. As a result, generation of soot due to combustion can be effectively suppressed compared to a case where the injection interval time is set fixedly.

Furthermore, the inventors of the present application have discovered that in an engine having an injector that has multiple injection holes and injects fuel radially from each injection hole, when the engine speed is high, compared to when the engine speed is low, We have obtained the knowledge that the swirl flow has a greater influence on fuel spray, and as a result, when the engine speed is high, the timing of oxygen arrival becomes earlier than when the engine speed is low. . In the present invention, in addition to the above points, as a control that takes this point into consideration, the reduction rate of the injection interval time with respect to an increase in engine speed is made larger when the engine speed is high than when it is low. This makes it possible to more reliably cause the fuel spray by after-injection to reach the specific position at a time when the oxygen concentration at the specific position becomes high.

Additionally, if the injection interval time is variable according to the engine speed as described above, the injection timing of after-injection can be advanced depending on the conditions compared to a case where the injection interval time is fixed. , the fuel efficiency of the engine can be improved. For example, if the injection interval time is set to be constant regardless of engine speed, the temperature in the combustion chamber will be sufficiently reduced so that the amount of soot generated does not become excessive regardless of whether the engine speed is high or low. It is necessary to start the after-injection after waiting for this to occur, that is, until a relatively long period of time has passed since the end of the main injection (the expansion stroke has progressed to some extent). This reduces the proportion of combustion energy based on after-injection that is used as work, leading to deterioration in fuel efficiency. On the other hand, when the injection interval time is made variable according to the engine speed as in the present invention, there is no need to uniformly delay the start time of after-injection as described above, and the after-injection can vary depending on the conditions. The injection timing can be advanced. Thereby, the rate at which combustion energy based on after-injection is converted into work can be increased as much as possible, and the fuel efficiency of the engine can be improved.

Here, it is known that the oxygen arrival timing also changes depending on various parameters other than the engine rotation speed. For example, even if the engine speed is the same, if the temperature of the combustion chamber is different, the timing of oxygen arrival changes depending on the difference. Specifically, when the temperature of the combustion chamber is high, the length of the fuel spray (spray length) is longer than when it is low, and this advances the timing of oxygen arrival. Therefore, the control device further includes a temperature sensor that detects a predetermined temperature parameter that increases as the engine warms up, and the injection control unit controls the Preferably , the higher the temperature parameter detected by the temperature sensor is, the shorter the injection interval time is.

According to this configuration, a good air utilization rate can be ensured without any problem regardless of the temperature of the combustion chamber, and the amount of soot generated can be effectively reduced.

As explained above, according to the diesel engine control device of the present invention, while maintaining relatively good fuel efficiency, the air utilization rate of the injected fuel is increased by after-injection, and soot generation is sufficiently suppressed. can do.

1 is a schematic system diagram showing a preferred embodiment of a diesel engine to which a control device of the present invention is applied. It is a figure which shows the structure of the crown surface of the piston in the said diesel engine, (a) is a perspective view, (b) is a sectional perspective view. FIG. 3 is a cross-sectional view for explaining the detailed structure of a cavity formed in the crown surface of the diesel piston and the flow of fuel spray injected into the cavity. FIG. 2 is a diagram for explaining the flow of fuel spray injected into the combustion chamber, and is a schematic plan view showing the crown surface of the piston together with other combustion chamber forming surfaces and the tip of the injector. FIG. 2 is a diagram for explaining the flow of fuel spray injected into the combustion chamber, and is a schematic plan view showing the crown surface of the piston together with other combustion chamber forming surfaces and the tip of the injector. It is a block diagram showing a control system of the above-mentioned diesel engine. It is an operation map showing the diffusion combustion region of the above-mentioned diesel engine. 3 is a time chart showing a fuel injection pattern adopted at a specific operating point within the diffusion combustion region. It is a flowchart which shows the specific procedure of fuel injection control performed in the said diffusion combustion region. FIG. 3 is a diagram schematically showing the flow of fuel spray flowing inside the cavity, where (a) shows the state of the spray at the end of main injection, and (b) and (c) show the state of the spray over time after the end of main injection. Each figure shows the state of the spray that has changed accordingly. It is a graph showing a time change in oxygen concentration at a specific position (turning reference point) in the cavity. It is a figure which shows typically the positional relationship between the fuel spray by the said main injection, and the fuel spray by the subsequent after-injection. It is a graph showing a time change in oxygen concentration at a specific position (swirling reference point) caused by vertical swirling flow. This is a group of graphs showing how the vertical swirling flow of fuel spray due to main injection changes depending on various parameters, in which (a) shows the relationship between the main injection amount, injection pressure, intake pressure, and swirling speed, and (b) (c) shows the relationship between injection amount, injection pressure, intake pressure, engine speed, and turning distance; (c) shows the relationship between main injection amount, injection pressure, intake pressure, engine speed, engine water temperature, and fuel temperature and spray length. Each shows the relationship. It is a graph showing a time change in oxygen concentration at a specific position (swirling reference point) caused by horizontal swirling flow. It is a graph showing the relationship between engine rotation speed and swirl speed. It is a graph showing the relationship between the elapsed time from the end of the main injection and the vertical turning weight coefficient. It is a graph showing the relationship between engine rotation speed and vertical turning weight coefficient. 3 is a group of graphs showing the relationship between the injection interval time, which is the time from the end of the main injection to the start of the after injection, and parameters of main injection amount, injection pressure, intake pressure, engine water temperature, and fuel temperature. It is a graph showing the relationship between the above-mentioned injection interval time and engine rotation speed. It is a graph showing the time change of the oxygen concentration at a specific position (turning reference point) in the cavity at different engine speeds. It is a graph showing the relationship between the injection interval time and engine rotation speed at different engine water temperatures. It is a time chart expressing the injection amount and injection interval time which change with the said acceleration driving|operation by the change of an injection waveform (injection pattern).

<Overall engine configuration>
FIG. 1 is a schematic system diagram showing a preferred embodiment of a diesel engine to which a control device of the present invention is applied. The diesel engine shown in this figure is a four-stroke diesel engine that is mounted on a vehicle as a driving power source. A diesel engine has an engine body 1 that is driven by receiving fuel mainly composed of light oil, an intake passage 30 through which intake air introduced into the engine body 1 flows, and an exhaust gas discharged from the engine body 1. It includes an exhaust passage 40 that circulates, an EGR device 44 that recirculates a part of the exhaust gas that flows through the exhaust passage 40 to the intake passage 30, and a turbo supercharger 36 that is driven by the exhaust gas that passes through the exhaust passage 40. ing.

The engine body 1 is of an in-line multi-cylinder type having a plurality of cylinders 2 (only one of which is shown in FIG. 1) arranged in a direction perpendicular to the paper surface of FIG. The engine body 1 includes a cylinder block 3 including a plurality of cylindrical cylinder liners defining a plurality of cylinders 2, and a cylinder head 4 attached to the upper surface of the cylinder block 3 so as to close the upper opening of each cylinder 2. , and a plurality of pistons 5 accommodated in each cylinder 2 so as to be able to reciprocate and slide. Note that since the structure of each cylinder 2 is the same, the following description will basically focus on only one cylinder 2.

A combustion chamber 6 is formed above the piston 5. The combustion chamber 6 is a space defined by the lower surface of the cylinder head 4 (combustion chamber ceiling surface 6U; see FIG. 3), the inner peripheral surface of the cylinder 2 (cylinder liner), and the crown surface 50 of the piston 5. . The combustion chamber 6 is supplied with the fuel by injection from an injector 15, which will be described later. The supplied mixture of fuel and air is combusted in the combustion chamber 6, and the piston 5, which is pushed down by the expansion force caused by the combustion, reciprocates in the vertical direction.

A crankshaft 7, which is an output shaft of the engine body 1, is provided below the piston 5. The crankshaft 7 is connected to the piston 5 via a connecting rod 8, and rotates around a central axis in accordance with the reciprocating motion (up and down motion) of the piston 5.

A crank angle sensor SN1 and a water temperature sensor SN2 are attached to the cylinder block 3. The crank angle sensor SN1 detects the rotation angle (crank angle) of the crankshaft 7 and the rotational speed (engine rotational speed) of the crankshaft 7. Water temperature sensor SN2 detects the temperature of cooling water (engine water temperature) flowing inside the cylinder block 3 and cylinder head 4. Note that this water temperature sensor SN2 corresponds to a "temperature sensor" in the claims , and the engine water temperature detected by the sensor corresponds to a "temperature parameter" in the claims .

The cylinder head 4 is formed with an intake port 9 and an exhaust port 10 that communicate with the combustion chamber 6 . An intake side opening, which is the downstream end of the intake port 9, and an exhaust side opening, which is the upstream end of the exhaust port 10, are formed on the lower surface of the cylinder head 4. The cylinder head 4 is assembled with an intake valve 11 that opens and closes an intake side opening and an exhaust valve 12 that opens and closes an exhaust side opening.

The cylinder head 4 is provided with an intake valve mechanism 13 and an exhaust valve mechanism 14 including a camshaft. The intake valve 11 and the exhaust valve 12 are driven to open and close by these valve mechanisms 13 and 14 in conjunction with the rotation of the crankshaft 7.

One injector 15 for injecting fuel into the combustion chamber 6 is attached to the cylinder head 4 for each cylinder 2 . The injector 15 has a tip 151 (FIG. 3) exposed at the ceiling of the combustion chamber 6, and the tip 151 is located on (or near) the cylinder axis X, which is the central axis of the cylinder 2. It is assembled to the cylinder head 4 in this manner. The injector 15 is capable of injecting fuel toward a cavity 5C (FIGS. 2 and 3), which will be described later, formed in the crown surface 50 of the piston 5.

A nozzle hole 152 (FIGS. 3 and 4), which serves as a fuel outlet, is formed at the tip 151 of the injector 15. The injector 15 is of a multi-nozzle hole type, and a plurality of nozzle holes 152 are formed in a tip portion 151 of the injector 15 . These plural nozzle holes 152 are arranged at equal pitches in the circumferential direction of the tip portion 151 of the injector 15 (FIG. 4). In this embodiment, ten nozzle holes 152 are formed in one injector 15. In plan view (viewed from the direction along the central axis X of the cylinder 2), the central axis AX of each nozzle hole 152 extends radially from the center of the tip portion 151 of the injector 15, that is, a point on the central axis X of the cylinder 2. The angles between the central axes AX of adjacent nozzle holes 152 are all the same. Further, the central axis AX of each nozzle hole 152 is inclined so that the further outward in the radial direction, the lower the center axis AX is located (FIG. 3). The fuel injected through such a nozzle hole 152 is radially injected from the tip 151 of the injector 15 toward the radially outer side and diagonally downward.

The injector 15 of each cylinder 2 is connected to a common rail 18 (accumulation rail) common to all cylinders 2 via a fuel supply pipe 17. High-pressure fuel pressurized by a fuel pump (not shown) is stored in the common rail 18 . The fuel accumulated in the common rail 18 is supplied to the injector 15 of each cylinder 2, so that the fuel is injected into the combustion chamber 6 from each injector 15 at a high pressure (for example, about 150 MPa to 250 MPa).

The injector 15 is provided with an injection pressure sensor SN5 (FIG. 6) that detects the pressure of the fuel inside the injector 15, in other words, the injection pressure that is the pressure of the fuel injected from the injector 15. One injection pressure sensor SN5 is provided for each of the plurality of injectors 15 corresponding to the plurality of cylinders 2.

Although not shown in FIG. 1, a fuel pressure regulator 16 and a fuel temperature sensor SN6 (see FIG. 6) are provided in a pipe connecting the fuel pump and the common rail 18. The fuel pressure regulator 16 adjusts the pressure of the common rail 18, that is, the pressure of fuel (fuel pressure) supplied to the injector 15. The fuel temperature sensor SN6 is a sensor that detects the temperature of fuel (fuel temperature) supplied to the injector 15.

The turbocharger 36 includes a compressor 37 disposed in the intake passage 30, a turbine 38 disposed in the exhaust passage 40, and a turbine shaft 39 connecting the compressor 37 and the turbine 38. The turbine 38 rotates by receiving the energy of the exhaust gas flowing through the exhaust passage 40. The compressor 37 compresses (supercharges) the air flowing through the intake passage 30 by rotating in conjunction with the rotation of the turbine 38 .

The intake passage 30 is connected to one side of the cylinder head 4 so as to communicate with the intake port 9. Air (fresh air) taken in from the upstream end of the intake passage 30 is introduced into the combustion chamber 6 through the intake passage 30 and the intake port 9. In the intake passage 30, an air cleaner 31, a compressor 37, a throttle valve 32, an intercooler 33, and a surge tank 34 are arranged in order from the upstream side.

The air cleaner 31 cleans the intake air by removing foreign matter from the intake air. The throttle valve 32 is an electrically operated butterfly valve that can adjust the flow rate of intake air in the intake passage 30. The compressor 37 is an impeller that compresses intake air and sends it out downstream of the intake passage 30. The intercooler 33 is a heat exchanger that cools intake air compressed by the turbocharger 36 (compressor 37). The surge tank 34 is a tank that provides a space for evenly distributing intake air to the plurality of cylinders 2, and is disposed immediately upstream of the intake manifold connected to the intake port 9 of each cylinder 2.

In the intake passage 30, an air flow sensor SN3 and an intake pressure sensor SN4 are arranged. The air flow sensor SN3 is arranged downstream of the air cleaner 31 and detects the flow rate of intake air passing through this part. The intake pressure sensor SN4 is arranged in the surge tank 34 and detects the pressure of intake air passing through the surge tank 34. In addition, since the surge tank 34 is disposed downstream of the compressor 37 of the turbocharger 36, the intake pressure detected by the intake pressure sensor SN4 indicates that the intake pressure has been supercharged by the turbocharger 36 (compressor 37). This is the later intake pressure, that is, the supercharging pressure.

The exhaust passage 40 is connected to the other side of the cylinder head 4 so as to communicate with the exhaust port 10. Burnt gas (exhaust gas) generated in the combustion chamber 6 is exhausted to the outside of the vehicle through the exhaust port 10 and the exhaust passage 40. In the exhaust passage, a turbine 38 and an exhaust purification device 41 are arranged in this order from the upstream side.

The turbine 38 is an impeller that rotates by receiving the energy of the exhaust gas, and applies rotational force to the compressor 37 in the intake passage 30 via the turbine shaft 39. The exhaust purification device 41 purifies harmful components in exhaust gas.

The exhaust purification device 41 includes an oxidation catalyst 42 that oxidizes CO and HC in the exhaust gas to render it harmless, and a DPF (diesel particulate filter) 43 that collects particulate matter contained in the exhaust gas. It has a built-in.

The EGR device 44 includes an EGR passage 45 that connects the exhaust passage 40 and the intake passage 30, and an EGR valve 46 that is provided in the EGR passage 45 and can be opened and closed. The EGR passage 45 connects a portion of the exhaust passage 40 upstream of the turbine 38 and a portion of the intake passage 30 between the intercooler 33 and the surge tank 34. The EGR valve 46 adjusts the flow rate of exhaust gas (EGR gas) that is returned from the exhaust passage 40 to the intake passage 30 through the EGR passage 45.

<Detailed structure of the piston>
Next, the structure of the piston 5, particularly the structure of the crown surface 50, will be described in detail. FIG. 2A is a perspective view mainly showing the upper part of the piston 5 (near the crown surface 50). FIG. 2(b) is a cross-sectional perspective view of the piston 5 shown in FIG. 2(a) taken along a vertical plane including the cylinder axis X. FIG. 3 is an enlarged sectional view showing a part of the crown surface 50 of the piston 5 together with other combustion chamber forming surfaces (inner peripheral surface of the cylinder 2 and combustion chamber ceiling surface 6U).

The piston 5 has the above-mentioned crown surface 50 that defines the bottom surface of the combustion chamber 6, and a cylindrical side peripheral surface 56 that continues to the outer peripheral edge of the crown surface 50.

A cavity 5C is formed in the crown surface 50 by recessing a main region including the center portion thereof downward (on the side opposite to the cylinder head 4). In other words, the crown surface 50 includes a wall surface defining the cavity 5C (a bottom portion 511, an outer peripheral portion 512, a lip portion 513, a shelf portion 521, and a rising portion 522, which will be described later) and an annular portion formed on the radially outer side of the cavity 5C. It has a squish surface 55 that is a flat surface.

The cavity 5C is a so-called reentrant type cavity. In particular, the cavity 5C of this embodiment is a reentrant type cavity with two upper and lower stages including a first cavity part 51 and a second cavity part 52. The first cavity portion 51 is a recess formed in a region including the radial center of the crown surface 50, and the second cavity portion 52 is an annular recess formed above the first cavity portion 51 in the crown surface 50. It is a recess.

The crown surface 50 has a bottom portion 511 , an outer peripheral portion 512 , and a lip portion 513 as a wall surface defining the first cavity portion 51 .

The bottom portion 511 is a wall portion that defines the bottom surface of the first cavity portion 51 . The bottom portion 511 is formed to have a gentle mountain shape, and has a top portion 511a at a radially central portion directly below the injector 15 (a position facing the tip portion 151 of the injector 15). That is, the bottom portion 511 is formed so that the height thereof gradually decreases from the top portion 511a toward the outside in the radial direction. The height of the bottom portion 511 is set to be lowest at the first boundary portion W1, which is the boundary between the bottom portion 511 and the outer peripheral portion 512.

The outer peripheral part 512 is a wall part connected to the radially outer side of the bottom part 511, and has a concave shape convex radially outwardly in a cross-sectional view. The outer peripheral part 512 has a concave shape so as to smoothly connect the first boundary part W1, which is the boundary between the bottom part 511 and the outer peripheral part 512, to the second boundary part W2, which is the boundary between the outer peripheral part 512 and the lip part 513. It is curved to. That is, the outer circumferential portion 512 includes a first portion that is curved so that the height gradually increases toward the outside in the radial direction from the first boundary portion W1, and a curved portion that is curved so that the height gradually increases from the first boundary portion W1 toward the second boundary portion W2 from the upper end of the first portion. The second portion is curved so that the diameter thereof gradually decreases. In other words, the outer peripheral portion 512 is formed so as to be recessed most radially outward at the intermediate portion M (FIG. 3) that is the boundary between the first and second portions.

The lip portion 513 is a wall portion continuous to the upper side of the outer peripheral portion 512, and has a shape that protrudes radially inward in a cross-sectional view. The lip portion 513 extends from a second boundary portion W2, which is a boundary between the outer peripheral portion 512 and the lip portion 513, to a boundary between the lip portion 513 and a shelf portion 521, which will be described later (in other words, a boundary between the first cavity portion 51 and the second cavity portion). It is curved in a convex shape (bump shape) so as to smoothly connect to the third boundary portion W3, which is the boundary with 52.

The crown surface 50 includes, in addition to the wall surfaces (bottom section 511, outer peripheral section 512, and lip section 513) that define the first cavity section 51 as described above, a shelf section 521 that is a wall surface that defines the second cavity section 52. and a rising portion 522.

The shelf portion 521 is a wall portion that defines the bottom surface of the second cavity portion 52 and is connected to the outside of the lip portion 513 of the first cavity portion 51 in the radial direction. The height of the shelf portion 521 gradually decreases from the third boundary portion W3, which is the boundary between the lip portion 513 and the shelf portion 521, to the fourth boundary portion W4, which is the boundary between the shelf portion 521 and the rising portion 522. It's slanted like that.

The rising portion 522 is a wall portion continuous to the outside in the radial direction of the shelf portion 521, and has a shape that rises upward from the shelf portion 521. The rising portion 522 is curved so as to smoothly connect the fourth boundary W4, which is the boundary between the shelf portion 521 and the rising portion 522, to the inner circumferential edge of the squish surface 55, and curves outward in the radial direction. It is formed so that the height gradually increases.

<Fuel spray flow>
Next, the flow of fuel spray injected from the injector 15 will be explained using FIGS. 3, 4, and 5. 4 and 5 are schematic plan views showing the crown surface 50 of the piston 5 together with another combustion chamber forming surface (inner peripheral surface of the cylinder 2) and the tip 151 of the injector 15. 3, 4 and 5 schematically show the distribution of gas within the combustion chamber 6. For example, in FIGS. 3 and 4, the symbol FS represents the spray of fuel immediately after the fuel is injected from the injector 15 while the piston 5 is located at or near compression top dead center. Further, in FIG. 3, the main flows of the fuel spray after the fuel spray FS collides with the wall surface (lip portion 513) of the cavity 5C are represented by symbols F11, F12, F13, F21, F22, and F23. In addition, when fuel is injected near compression top dead center in a diesel engine like this embodiment, the fuel starts to burn after a short time after injection (diffusion combustion). Therefore, the fuel spray FS basically contains combustion gas in addition to the atomized fuel. However, in this specification, fuel spray containing combustion gas and fuel spray not containing combustion gas are simply referred to as fuel spray (or spray) without any particular distinction.

Here, the flow of the fuel spray FS viewed from the direction orthogonal to the central axis X of the cylinder 2 and the flow of the fuel spray FS viewed from the direction along the central axis X of the cylinder 2 will be explained separately. The explanation will be made assuming that there is no influence of the other flow on the flow.

First, the flow of the fuel spray FS viewed from a direction perpendicular to the central axis X of the cylinder 2 will be described using FIG. 3. The fuel injected from the nozzle hole 152 of the injector 15 is atomized while being diffused at a spray angle θ, and flies along the injection axis AX. The injection axis AX is the main axis of the fuel spray FS and is an axis extending from the central axis of the nozzle hole 152. When the piston 5 is at or near compression top dead center, the fuel injected from the nozzle hole 152 (fuel spray FS) is directed toward the lip portion 513 of the cavity 5C. In other words, the injector 15 has a nozzle hole 152 that can direct fuel injected at or near compression top dead center to the lip portion 513.

The fuel spray FS injected toward the lip portion 513 collides with the lip portion 513, and then the spray is directed toward the first cavity portion 51 (downward) (arrow F11) and toward the second cavity portion 52 (arrow F11). and the spray (arrow F21) directed upward). The separated spray flows along the wall shapes of the first and second cavity parts 51 and 52 while being mixed with the air present in the first and second cavity parts 51 and 52, respectively.

Specifically, the spray indicated by the arrow F11 is directed downward at the lip portion 513 and enters the outer peripheral portion 512 of the first cavity portion 51. The spray that has entered the outer peripheral part 512 changes its flow direction from below to the inside in the radial direction along the curved shape of the outer peripheral part 512, and then flows along the wall shape of the bottom part 511 as shown by arrow F12. Since the bottom portion 511 is formed so as to rise toward the inside in the radial direction, the spray indicated by the arrow F12 is lifted upward, and finally changes direction so as to go outward and upward in the radial direction as indicated by the arrow F13. It flows so as to return to the position on the injection axis AX, which is the main axis of the initial spray (spray FS immediately after exiting from the injection hole 152). In this way, the spray that has entered the first cavity section 51 swirls and flows within the first cavity section 51 so as to form a vortex around the axis perpendicular to the central axis X of the cylinder 2. Hereinafter, the direction of rotation around an axis perpendicular to the central axis X of the cylinder 2 will be referred to as the longitudinal direction. Further, the vertical swirling flow of gas (fuel spray or air) is referred to as a vertical swirling flow, as appropriate.

On the other hand, the spray indicated by the arrow F21 is directed upward at the lip portion 513 and enters the shelf portion 521 of the second cavity portion 52. The spray that has entered the shelf section 521 flows diagonally downward along the inclination of the shelf section 521, and is then lifted upward along the curved wall surface of the rising section 522 as shown by arrow F22, and finally flows radially inward along the combustion chamber ceiling surface 6U.

Here, the upper end portion of the rising portion 522 is not provided with a shaped portion that protrudes inward in the radial direction like the lip portion 513. Therefore, the flow of the spray indicated by arrow F22 is not excessively strengthened, and spray (arrow F23) that branches from arrow F22 and flows radially outward is also generated. In particular, in the later stages of combustion, the fuel may be pulled by a reverse squish flow (a flow directed from the inside to the outside in the radial direction along the squish surface 55), and the flow shown by arrow F23 is likely to occur. This promotes the use of the air present above the squish surface 55, which leads to suppressing the generation of soot.

As the spray that has entered the second cavity portion 52 as described above branches into two directions indicated by arrows F22 and F23, the spray is dispersed over a relatively wide range in the upper part of the combustion chamber 6. Therefore, the flow of each spray after branching is not so strong, and in particular, the flow of arrow F22 after changing direction inward in the radial direction is relatively weak. Due to these circumstances, the spray indicated by arrow F22 does not substantially generate a swirling flow that returns to the position on the injection axis AX. In this respect, it is different from the spray inside the first cavity part 51, which generates swirling flows as shown by arrows F11, F12, and F13.

Next, the flow of the fuel spray FS viewed from the direction along the central axis X of the cylinder 2 will be explained using FIGS. 4 and 5.

As described above, ten nozzle holes 152 are formed at equal pitches in the tip 151 of the injector 15, and the fuel injected from each nozzle hole 152 flies along each injection axis AX. From this, immediately after fuel is injected from the injector 15, as shown in FIG. 4, when viewed from the direction along the central axis X of the cylinder 2, the fuel spray FS is concentrated around each injection axis AX, and The gas in the region between the adjacent injection axes AX becomes gas AS whose main component is air. When a swirl flow is not generated in the combustion chamber 6, each fuel spray FS remains around each injection axis AX for a while after the fuel is injected from the injector 15, and then diffuses. On the other hand, if a swirl flow is generated within the combustion chamber 6, that is, a gas flow around the central axis X of the cylinder 2 is generated within the combustion chamber 6 as shown by the arrow Y1 in FIGS. 4 and 5. In this case, as shown in FIG. 5, each fuel spray FS is caused to flow in the circumferential direction of the cylinder 2 by a swirl flow, and turns around the central axis X of the cylinder 2. From this, fuel spray FS and clean gas AS (gas whose main component is air) alternately flow into the region on the injection axis AX. Note that FIG. 5 schematically shows the main part of the fuel spray FS, and in reality, the fuel spray FS turns around the central axis X of the cylinder 2 while diffusing to the surroundings. Hereinafter, the direction of rotation of the cylinder 2 around the central axis X will be referred to as the lateral direction. Further, the horizontal swirling flow of gas (fuel spray or air) is referred to as a transverse swirling flow, as appropriate.

<Control system>
FIG. 6 is a block diagram showing the control system of the diesel engine. The ECU 70 shown in this figure is a microprocessor for controlling the engine in an integrated manner, and is composed of a well-known CPU, ROM, RAM, and the like. This ECU 70 corresponds to the "injection control section" in the claims.

Detection information from various sensors is input to the ECU 70. For example, the ECU 70 is electrically connected to the above-described crank angle sensor SN1, water temperature sensor SN2, air flow sensor SN3, intake pressure sensor SN4, injection pressure sensor SN5, and fuel temperature sensor SN6. Information detected by these sensors SN1 to SN6, such as crank angle, engine speed, engine water temperature, intake air amount, intake pressure, fuel injection pressure, and fuel temperature, is sequentially input to the ECU 70.

The vehicle is also provided with an accelerator opening sensor SN7 that detects the opening of an accelerator pedal operated by the driver of the vehicle. Information detected by the accelerator opening sensor SN7 is also sequentially input to the ECU 70.

The ECU 70 controls each part of the engine while executing various determinations and calculations based on information inputted from each of the sensors SN1 to SN7. That is, the ECU 70 is electrically connected to the injector 15, fuel pressure regulator 16, throttle valve 32, EGR valve 46, etc., and sends control signals to each of these devices based on the results of the above-mentioned determination and calculation. Output.

<Fuel injection control in the diffusion combustion region>
Next, as a typical fuel injection control in the above engine, injection control in the diffusion combustion region A1 shown in FIG. 7 will be described. The diffusion combustion region A1 shown in FIG. 7 is an operating region in which most of the fuel injected from the injector 15 is burned by diffusion combustion, and is the main operating region excluding the extremely low load region, extremely high load region, and extremely high speed region of the engine. is set in the area. Diffusion combustion, as is well known, is a combustion form widely adopted in diesel engines, in which the fuel injected from the injector 15 is evaporated and mixed with air due to the diffusion effect, resulting in a combustible part (the main part). This is a type of combustion in which the air-fuel mixture is combusted by self-ignition from the boundary between the fuel spray and the air).

When the total amount of fuel to be supplied from the injector 15 to the combustion chamber 6 (cylinder 2) during one combustion cycle is defined as the total injection amount, in the diffusion combustion region A1 of FIG. An injection pattern is adopted in which fuel is injected at or near compression top dead center. FIG. 8 is a time chart showing the injection pattern adopted at a typical operating point C1 within the diffuse combustion region A1, in which the horizontal axis is the crank angle (deg) and the vertical axis is the fuel injection rate (deg) based on the crank angle. mm ³ /deg).

As shown in FIG. 8 , in the diffusion combustion region A1, three pre-injections JP, one main injection Jm, and one after-injection Ja are performed. The main injection Jm is a fuel injection performed at or near the compression top dead center (TDC), which is the top dead center (TDC) between the compression stroke and the expansion stroke. Executed over a period of time. The injection period of the main injection Jm includes at least the timing at which the injection axis AX of the injector 15 and the lip portion 513 of the first cavity portion 51 intersect (see FIG. 3). Pre-injection Jp is fuel injection performed during the compression stroke before main injection Jm. The after injection Ja is a fuel injection performed during the expansion stroke after the main injection Jm. Among the main, pre, and after injections, the main injection Jm injects the largest proportion of fuel out of the total injection amount during one combustion cycle.

Here, in this embodiment, when the total injection amount is increased as the engine load increases, the increase in fuel is mainly allocated to the main injection Jm.

The fuel injection pattern in the diffusion combustion region A1 is basically determined with reference to predetermined map data. Specifically, the storage unit of the ECU 70 stores the injection amount and injection timing (or number of injections) of pre-injection JP, the injection amount and injection timing of main injection Jm, and the injection amount of after-injection Ja based on operating conditions (load map data determined for each engine (speed, rotation speed, etc.) is stored in advance. When operating in the diffuse combustion region A1, the ECU 70 refers to the map data to determine an injection pattern that suits the operating conditions (operating points) at the time, and injects fuel from the injector 15 according to the determined injection pattern. let

However, the injection timing of after-injection Ja is determined each time by calculation without using map data. Although the details will be described later, the injection timing of after injection Ja is a desirable time when the injection interval time (Ti in Fig. 8), which is the time from the end of main injection Jm to the start of after injection Ja, is determined from the viewpoint of air utilization rate. It is determined that

Further, the storage unit of the ECU 70 stores map data in which a target injection pressure, which is a target value of the injection pressure of fuel from the injector 15, is predetermined for each engine operating condition (load, rotation speed, etc.). The fuel pressure regulator 16 is controlled so that the actual injection pressure matches the target injection pressure. The target injection pressure is set to increase as the engine load increases and the total injection amount during one combustion cycle increases. This is because by increasing the amount of fuel that can be injected per unit time, a relatively large amount of fuel commensurate with the high load can be injected within a limited time. Conversely, when the engine load is low (the total injection amount is small), the target injection pressure is lowered, so the load on the fuel pump can be reduced and fuel efficiency can be improved.

Next, the procedure of fuel injection control in the diffusion combustion region A1 will be explained based on the flowchart of FIG. When the control shown in the flowchart starts, the ECU 70 determines whether the current operating point of the engine is included in the diffusion combustion region A1 shown in FIG. 7 (step S1). That is, the ECU 70 determines the current engine speed based on the engine speed detected by the crank angle sensor SN1 and the engine load (required torque) specified from the detected value (accelerator opening) of the accelerator opening sensor SN7. It is determined whether or not the operating point is included in the diffusion combustion region A1.

If the determination in step S1 is YES and it is confirmed that the current operating point is included in the diffusion combustion region A1, the ECU 70 controls the total injection amount, which is the total amount of fuel to be injected from the injector 15 during the next one combustion cycle. The amount and the injection pattern for injecting fuel corresponding to the total injection amount are determined (step S2). For example, the total injection amount is determined to increase as the engine load increases, and the injection pattern is determined based on the above-mentioned map data stored in advance in the storage section of the ECU 70. The injection pattern determined here includes the injection amount and injection timing (or number of injections) of the pre-injection Jp, the injection amount and injection timing of the main injection Jm, and the injection amount of the after-injection Ja. On the other hand, the injection timing of the after injection Ja is not determined here, but is determined based on the injection interval time calculated in step S6, which will be described later.

Next, the ECU 70 determines whether the closing timing (IVC) of the intake valve 11 has arrived (step S3). That is, the ECU 70 determines whether or not the intake valve 11 in the cylinder 2 to which fuel is to be injected is closed.

If the determination in step S3 is YES and it is confirmed that the closing timing of the intake valve 11 has arrived, the ECU 70 detects the fuel injection pressure, intake pressure, engine speed, engine water temperature, and fuel temperature from each sensor. Acquire (step S4). Specifically, the ECU 70 obtains the fuel injection pressure from the detection value of the injection pressure sensor SN5 provided in the injector 15 of the target cylinder 2, obtains the intake pressure from the detection value of the intake pressure sensor SN4, and obtains the intake pressure from the detection value of the intake pressure sensor SN4. The engine rotation speed is obtained from the detected value of SN1, the engine water temperature is obtained from the detected value of water temperature sensor SN2, and the fuel temperature is obtained from the detected value of fuel temperature sensor SN6.

Next, the ECU 70 uses the information acquired in step S4 (injection pressure, intake pressure, engine speed, engine water temperature, fuel temperature) and the injection amount of main injection Jm determined in step S2. , a vertical rotation frequency f1, an initial phase φ, a horizontal rotation frequency f2, and a vertical rotation weighting coefficient α, which will be described later, are parameters that define the oxygen concentration fluctuation waveform at the rotation reference point Z (step S5).

The swirl reference point Z is a point on the injection axis AX (an extension of the central axis of the nozzle hole 152) as shown in FIG. This is the intersection of the main axis of AX and the injection axis AX.

Specifically, FIG. 10 is a diagram schematically showing the flow of the fuel spray caused by the main injection Jm that vertically swirls inside the first cavity part 51, and (a) shows the state of the spray at the end of the main injection Jm. , (b) and (c) respectively show the state of the spray that changed with the passage of time after the end of the main injection Jm. As shown in FIG. 10, a spray Fm of fuel injected by the main injection Jm near the compression top dead center (or over a predetermined period beyond the compression top dead center) (actually, it is a combination of combustion gas and atomized fuel). ) swirls to form a vertical vortex along each wall surface of the lip part 513 , outer peripheral part 512 , and bottom part 511 that constitute the first cavity part 51 , and the injection axis AX of the injector 15 ( It returns to the position above the extension line of the central axis of the nozzle hole 152. The swirl reference point Z is the intersection of the main axis of the fuel spray Fm that has returned onto the injection axis AX due to the vertical swirl flow and the injection axis AX.

The oxygen concentration at the swirl reference point Z varies depending on the degree of progress of the vertical swirl flow of the fuel spray Fm. That is, at the time of FIG. 10A, which is the end of main injection Jm, the fuel spray Fm is passing over the turning reference point Z, so the oxygen concentration at the turning reference point Z becomes very thin. This state of low oxygen concentration continues until the time point in FIG. 10(b) when the fuel spray Fm has completely passed through the turning reference point Z. However, at this point, as shown by the white arrow E, a flow of air with a high oxygen content (hereinafter referred to as a clean air flow) is generated as it is attracted to the negative pressure generated at the rear end of the fuel spray Fm. This clean air flow E starts flowing into the turning reference point Z. As a result, the oxygen concentration at the turning reference point Z gradually increases after the time shown in FIG. 10(b). Thereafter, at the point in time in FIG. 10(c) when the rear end of the fuel spray Fm has left the turning reference point Z, a state is obtained in which the middle part of the clean air flow E passes through the turning reference point Z, and at this point, the turning reference point Z is reached. The oxygen concentration at point Z is the highest. Note that the oxygen concentration gradually increases between FIG. This is because the closer to the center of the direction, the higher the height.

The oxygen concentration at the swirling reference point Z also varies depending on the horizontal swirling flow of the fuel spray Fm. That is, the fuel spray Fm injected by the main injection Jm is swept away by the swirl flow as described above and turns around the central axis X of the cylinder 2, and on the injection axis AX there is a gas containing a large amount of the fuel spray FS. A gas whose main component is air enters alternately. In this way, the oxygen concentration at a point on the injection axis AX that includes the swirl reference point Z varies also due to the horizontal swirl flow.

In step S5 described above, the ECU 70 determines the vertical swirl frequency f1 and initial phase φ that define the fluctuation waveform of oxygen concentration at the swirl reference point Z associated with the vertical swirl flow, and the oxygen concentration at the swirl reference point Z associated with the horizontal swirl flow. The horizontal rotation frequency f2 that defines the fluctuation waveform of is calculated. Furthermore, in step S5, the ECU 70 calculates a vertical rotation weighting coefficient α that represents the degree of influence of the vertical rotation flow and the horizontal rotation flow on the oxygen concentration at the rotation reference point Z.

The details of the procedure for calculating the vertical rotation frequency f1, initial phase φ, horizontal rotation frequency f2, and vertical rotation weighting coefficient α will be described later, but for example, the fluctuation waveform of the oxygen concentration at the rotation reference point Z is the line L1 shown by the solid line in FIG. The oxygen concentration at the turning reference point Z reaches a maximum at a predetermined time t11. In the graph of FIG. 11, the local λ, which is the value obtained by dividing the local air-fuel ratio at the turning reference point Z by the stoichiometric air-fuel ratio, is used as a parameter representing the oxygen concentration at the turning reference point Z, and the value of this local λ is used as a parameter representing the oxygen concentration at the turning reference point Z. is taken on the vertical axis. The larger the local λ, the higher the oxygen concentration. Further, t on the horizontal axis is the time change (msec) from the end of the main injection Jm.

Next, the ECU 70 generates an oxygen concentration fluctuation waveform (vertical rotation frequency f1 , the initial phase φ, the oxygen concentration at the turning reference point Z at each point in time calculated using the horizontal turning frequency f2 and the vertical turning weighting coefficient α), the time from the end of the main injection Jm to the start of the after injection Ja. An injection interval time Ti is determined (step S6). This injection interval time Ti is a time such that the tip of the fuel spray Fa caused by the after injection Ja reaches the turning reference point Z at the time when the oxygen concentration at the turning reference point Z becomes the highest (time t11 in the example of FIG. 11). is set to Hereinafter, the time when the oxygen concentration at the turning reference point Z is the highest will be referred to as the oxygen arrival time. This oxygen arrival timing corresponds to when the middle part of the clean air flow E passes through the turning reference point Z as shown in FIG. 10(c) or FIG. 12. Using this to rephrase the injection interval time Ti, the injection interval time Ti is determined by determining the time when the tip of the fuel spray Fa due to the after-injection Ja reaches the turning reference point Z, and the time when the middle part of the clean air flow E is turning. The time is set to match the time when the reference point Z is reached.

Next, the ECU 70 causes the injector 15 to perform pre-injection Jp and main injection Jm (step S7). Note that the pre-injection Jp and the main injection Jm here are executed according to the injection pattern determined based on the predetermined map data in step S2 (the injection pattern that determines the injection amount and injection timing of each pre-main injection). be done.

Next, the ECU 70 determines whether the elapsed time since the end of the main injection Jm executed in step S7 has reached the injection interval time Ti determined in step S6 (step S8).

If the determination in step S8 is YES and it is confirmed that the injection interval time Ti has elapsed since the end of the main injection Jm, the ECU 70 causes the injector 15 to start the after injection Ja at that time (step S9). As a result, as shown in FIG. 12, the time when the tip of the fuel spray Fa caused by the after-injection Ja reaches the turning reference point Z on the injection axis AX is determined as the oxygen arrival time when the oxygen concentration at the turning reference point Z becomes high. Can be matched. This leads to increasing the air utilization rate when the fuel injected by after-injection Ja is combusted. Note that the injection amount determined in step S2 based on the predetermined map data is adopted as the injection amount of after-injection Ja in step S9.

<Changes in oxygen concentration at turning reference point Z>
In this embodiment, the fluctuation waveform of the oxygen concentration at the swirling reference point Z caused only by the vertical swirling flow and the fluctuating waveform of the oxygen concentration at the swirling reference point Z caused only by the horizontal swirling flow are obtained separately, and these are combined. The waveform is calculated as the oxygen concentration fluctuation waveform at the turning reference point Z. In the following, the waveform of fluctuation in oxygen concentration at the turning reference point Z caused only by the vertical turning flow will be referred to as the vertical oxygen concentration waveform, and the waveform of the fluctuation in the oxygen concentration at the turning reference point Z caused only by the horizontal turning flow will be referred to as the horizontal oxygen concentration waveform. That's what it means.

<Vertical oxygen concentration waveform>
FIG. 13 is a graph showing an example of a vertical oxygen concentration waveform. Similar to the graph in FIG. 10, in the graph in FIG. 11, the horizontal axis is the time t (msec) from the end of the main injection Jm, and the vertical axis is the local λ at the turning reference point Z.

As shown in FIGS. 10(a) to (c), the fuel spray Fm swirls and a clean air flow E is formed at the rear end of the fuel spray Fm, as shown by the solid line in the graph of FIG. The local λ at the turning reference point Z becomes larger as time passes from the end of the main injection Jm (t=0), and changes periodically such that it reaches a maximum value and then decreases again. Specifically, the local λ is very small (point Ra) at the time when the main injection Jm ends (t=0), and starts to rise significantly after the subsequent time t21 (point Rb). Furthermore, the local λ takes a maximum value at time t22 (point Rc), which is later than time t21, and gradually decreases thereafter. In this case, point Ra when t=0 corresponds to the state in FIG. 10(a), point Rb when t=t21 corresponds to the state in FIG. 10(b), and point Rb when t=t22 corresponds to the state in FIG. 10(b). Point Rc corresponds to the state shown in FIG. 10(c).

In this way, as the fuel spray Fm vertically swirls within the first cavity portion 51, the local λ (or oxygen concentration) at the swirl reference point Z periodically fluctuates. On the premise of this phenomenon, in this embodiment, the time change in oxygen concentration at the swirling reference point Z accompanying the vertical swirling flow, that is, the longitudinal oxygen concentration waveform, is represented by a periodic function.

Assuming that a periodic function representing the longitudinal oxygen concentration waveform is x1(t), this x1(t) is schematically defined by the following equation (1).

[Number 1]
x1(t)=cos(2π・f1・t−φ) ‥‥(1)
Here, f1 is the vertical rotation frequency, and φ is the initial phase. The vertical swirl frequency f1 is the number of swirls per unit time of the fuel spray that swirls in the vertical direction.

Further, as shown in FIG. 10(a), the length of the fuel spray Fm is defined as a spray length L, and the speed at which the fuel spray Fm flows (swivels) in the vertical direction is defined as a vertical swirl speed V1. Further, as shown in FIG. 10(b), the distance traveled from the turning reference point Z to the point of return to the turning reference point Z by following the broken line route (distance of the broken line route) is defined as the vertical turning distance D. The vertical swirl frequency f1 and the initial phase φ can be expressed by the following equation (2) using the spray length L, the vertical swirl speed V1, and the vertical swirl distance D, respectively.

[Number 2]
f1=V1/D
φ=2π×L/D (2)
That is, the vertical swirl frequency f1 is equal to the vertical swirl speed V1 divided by the swirl distance D, and the initial phase φ is equal to a constant times the spray length L divided by the swirl distance D.

From the above equation (2), in order to obtain the vertical swirl frequency f1 and the initial phase φ, it is necessary to know the vertical swirl speed V1, the swirl distance D, and the spray length L. According to the findings of the inventor of the present application, these values (V, D, L) are calculated based on the main injection amount (injection amount of main injection Jm), injection pressure, intake pressure, engine It can be expressed as a function of multiple parameters selected from rotational speed, engine water temperature, and fuel temperature.

[Number 3]
V1=F1 (main injection amount, injection pressure, intake pressure)
D=F2 (main injection amount, injection pressure, intake pressure, rotation speed)
L=F3 (main injection amount, injection pressure, intake pressure, rotation speed, water temperature, fuel temperature) (3)
In other words, the vertical turning speed V1 is a function using the main injection amount, injection pressure, and intake pressure as parameters (variables), and the turning distance D is a function using the main injection amount, injection pressure, intake pressure, and engine speed as parameters. The spray length L is a function whose parameters are the main injection amount, injection pressure, intake pressure, engine speed, engine water temperature, and fuel temperature.

FIG. 14 is a group of graphs for explaining the outline of the function of formula (3) above, in which (a) shows the relationship between the turning speed V and each parameter, and (b) shows the relationship between the turning distance D and each parameter. (c) shows the relationship between the spray length L and each parameter.

As shown in FIG. 14(a), the vertical turning speed V1 becomes faster as the injection amount of the main injection Jm is larger, becomes faster as the fuel injection pressure is higher, and becomes slower as the intake pressure is higher. As shown in FIG. 14(b), the turning distance D increases as the injection amount of the main injection Jm increases, increases as the fuel injection pressure increases, decreases as the intake pressure increases, and increases as the engine speed increases. It gets longer. As shown in FIG. 14(c), the spray length L increases as the injection amount of the main injection Jm increases, increases as the fuel injection pressure increases, decreases as the intake pressure increases, and decreases as the engine speed increases. The higher the engine water temperature, the longer it becomes; and the higher the fuel temperature, the longer it becomes. In addition, each graph of FIG. 14(a), (b), and (c) shows the changes in V1, D, and L obtained when the parameter shown on the horizontal axis changes alone (when other parameters are constant). shall be shown. Further, each graph is a linear graph representing a simple direct proportional or inverse proportional relationship, but it is only a schematic and is not necessarily a linear graph.

In step S5, the vertical rotation frequency f1 and the initial phase φ are calculated by a predetermined calculation using the above knowledge. That is, in the above step S5, the ECU 70 uses each information of the injection pressure, intake pressure, engine speed, engine water temperature, and fuel temperature acquired in the above step S4 and the main injection Jm determined in the above step S2. By substituting the amount into a pre-stored arithmetic expression corresponding to the above equation (3) (or FIG. 14), the vertical turning speed V1, the turning distance D, and the spray length L are calculated. Then, by substituting these calculated values (V1, D, L) into the pre-stored arithmetic equation corresponding to the above equation (2), the vertical rotation frequency f1 (=V1/D) and the initial phase are determined. Calculate φ (=2π×L/D). By substituting the vertical rotation frequency f1 and initial phase φ calculated in this way into equation (1), a vertical oxygen concentration waveform (function x1(t)) as shown in FIG. 13 is defined.

<Horizontal oxygen concentration waveform>
FIG. 15 is a graph showing an example of a lateral oxygen concentration waveform. Similar to the graphs in FIGS. 11 and 13, in the graph in FIG. 15, the horizontal axis is the time t (msec) from the end of the main injection Jm, and the vertical axis is the local λ at the turning reference point Z.

As mentioned above, gas containing a large amount of fuel spray FS and clean gas AS (gas whose main component is air) alternately enter on the injection axis AX due to the swirl flow, so that the solid line in the graph of FIG. As shown, the local λ at the turning reference point Z increases as time passes from the end of the main injection Jm (t=0), and then decreases again after reaching the maximum value, and so on periodically. Change. Specifically, the local λ is very small (point Rd1) at the time when main injection Jm ends (t=0), and when it gradually increases from that time and reaches its maximum value at time t31 (point Re1), It gradually decreases to the minimum value (point Rd2), rises again to the maximum value (Re2), and then gradually decreases again.

In this embodiment, the temporal change in oxygen concentration at the turning reference point Z due to this lateral turning flow, that is, the lateral oxygen concentration waveform, is also expressed by a periodic function. Assuming that the periodic function representing the lateral oxygen concentration waveform is x2(t), this x2(t) is schematically defined by the following equation (4).

[Number 4]
x2(t)=cos(2π・f2・t) (4)
Here, f2 is the horizontal rotation frequency. The horizontal swirl frequency f2 is the number of swirls per unit time of the fuel spray that swirls around the central axis It refers to the number of turns of the swirling flow in the direction.

Furthermore, as shown in FIGS. 4 and 5, the distance between adjacent turning reference points Z when viewed from the direction along the central axis X of the cylinder 2, that is, the distance between the turning reference points Z and the central axis Assuming that the distance between the injection axes AX on the circumference centered on is d, and the velocity of the swirl flow is VS, the horizontal swirl frequency f2 can be expressed using the following formula (5). I can do it.

[Number 5]
f2=VS/d (5)
That is, the transverse swirl frequency f2 is equal to the value obtained by dividing the velocity VS of the swirl flow by the distance d between the injection axes.

The distance between the injection axes d changes depending on the bore diameter of the cylinder 2, the number of nozzle holes 152, and the spray angle θ (see FIG. 4), but these values are predetermined for each engine, and the ECU 70 has these values. The distance d between injection axes calculated in advance from the value is stored as a constant. On the other hand, the speed VS of the swirl flow changes depending on the engine rotation speed. Specifically, as shown in FIG. 16, the higher the engine speed, the higher the speed VS of the swirl flow. Note that, like the graphs in FIG. 14, the graphs in FIG. 16 are only schematic, and the graphs representing the relationship between the engine rotation speed and the speed of the swirl flow VS are not necessarily linear graphs. .

In step S5, the lateral rotation frequency f2 is calculated by a predetermined calculation using the above knowledge. That is, in step S5, the ECU 70 converts the engine speed obtained in step S4 and the pre-stored distance d between injection axes into a pre-stored arithmetic expression corresponding to the above formula (5). By substituting, the horizontal rotation frequency f2 (=VS/d) is calculated. By substituting the lateral rotation frequency f2 calculated in this way into equation (4), a lateral oxygen concentration waveform (function x2(t)) as shown in FIG. 15 is defined.

<Fluctuation waveform of turning reference point Z>
Combining the vertical oxygen concentration waveform (function x1(t)) expressed by the above equation (1) and the horizontal oxygen concentration waveform (function x2(t)) expressed by equation (4), the turning reference point It is considered that the fluctuation waveform of the oxygen concentration in Z is obtained. However, the magnitude of the influence of the swirling reference point Z on the oxygen concentration is not the same between the vertical swirling flow and the horizontal swirling flow. From this, the fluctuation waveform of the oxygen concentration at the swirling reference point Z caused by the swirling flow in both the vertical and horizontal directions is determined by taking into account the degree of these influences. Specifically, a function x3(t ) is calculated using the following equation (5). The vertical rotation weighting coefficient α mentioned above is this ratio α.

[Formula 5]
x3(t)=α・x1(t)+x2(t) (5)
By substituting the above equations (1) to (4) into this equation (5), the function x3(t) is expressed as the following equation (6).

[Formula 6]
x3(t)=α・cos(2π・V1/D・t−2π×L/D)
+cos(2π・VS/d・t) (6)
The weaker the vertical swirling flow, the greater the influence of the horizontal swirling flow on the oxygen concentration at the swirling reference point Z, and the smaller the vertical swirling weighting coefficient α. The vertical swirling flow becomes weaker as the time elapses from the end of the main injection Jm. Therefore, as shown in FIG. 17, the vertical rotation weighting coefficient α is set to a smaller value as the elapsed time from the end of the main injection Jm becomes longer.

Furthermore, as the engine speed is higher, the speed VS of the swirl flow becomes higher and the lateral swirling flow becomes stronger, and the influence of the lateral swirling flow on the oxygen concentration at the swirling reference point Z becomes larger. However, even when the engine speed is high, when the penetration force of the fuel spray Fm is strong due to the high injection amount and injection pressure of the main injection Jm, the vertical swirl flow becomes strong and the lateral The influence of the swirling flow in the direction on the oxygen concentration at the swirling reference point Z is suppressed to a small extent.

From this, as shown by the solid line in FIG. 18, when the penetration force of the fuel spray Fm is weak, the vertical rotation weighting coefficient α is set to a smaller value as the engine speed is higher. Specifically, the vertical turning weighting coefficient α when the engine rotational speed is equal to or higher than the predetermined reference rotational speed N0 is set to a smaller value than the vertical turning weighting coefficient α when the engine rotational speed is less than the reference rotational speed N0. However, in the engine rotation speed range above the reference rotation speed N0, the vertical turning weighting coefficient α is a constant value regardless of the engine rotation speed, and even in the engine rotation speed range below the reference rotation speed N0, the vertical turning weighting coefficient α is set to a constant value regardless of the engine rotation speed. It is assumed to be a constant value regardless of the number.

Further, in the engine speed range below the reference speed N0, the vertical rotation weighting coefficient α is set to a constant value regardless of the penetration force of the fuel spray Fm. On the other hand, in the engine speed range above the reference speed N0, the vertical rotation weighting coefficient α is within the range of the value when the engine speed is below the reference speed N0, and the value increases as the penetration force of the fuel spray Fm becomes stronger. It is said that As shown in the dashed line graph in FIG. 18, when the penetration force of the fuel spray Fm becomes stronger than a predetermined value, the vertical rotation weighting coefficient α becomes greater than or equal to the reference rotation speed N0. It is assumed to be a constant value regardless of the number.

As described above, in the diffusion combustion region A1, the higher the engine load, the higher the main injection amount and injection pressure, and the greater the penetration force of the fuel spray Fm. From this, in the region A1_L where the engine speed is low in the diffusion combustion region A1 (region below the reference speed N0, hereinafter sometimes referred to as the low speed region), the vertical turning weighting coefficient α is a constant value regardless of the engine load. It is said that In addition, in the region where the engine speed is high in the diffusion combustion region A1 (the region where the engine speed is higher than the reference speed N0, hereinafter referred to as the high speed region), the higher the engine load, the larger the value of the vertical turning weighting coefficient α. . Here, in this embodiment, in the region A1_m (FIG. 7, hereinafter referred to as the diffusion maximum load region) where the engine load is maximum in the diffusion combustion region A1, vertical turning is performed in the high speed region A1_H and the low speed region A1_L. The weighting coefficient α is set to the same value, and in the remaining regions of the diffusion combustion region A1 excluding this region, the vertical rotation weighting coefficient α of the high speed region A1_H is set to a smaller value than the vertical rotation weighting coefficient α of the low speed region A1_L. Ru.

In the above step S5, the relationship between the elapsed time from the end of the main injection Jm, the engine speed, the penetration force of the fuel spray Fm, and the vertical rotation weighting coefficient α is determined as shown in the graphs of FIGS. 17 and 18. The vertical turning weighting coefficient α is calculated as follows.

Specifically, a map corresponding to the graph of FIG. 17 and serving as a reference for the elapsed time and vertical turning weighting coefficient α is set in advance and stored in the ECU 70. Further, a map in which the vertical axis of the graph in FIG. 18 is converted into a correction coefficient of the vertical turning weight coefficient α, and the penetration force in the graph in FIG. 18 is converted into engine load is set in advance and stored in the ECU 70. In step S5, the ECU 70 extracts a correction coefficient for the vertical turning weighting coefficient α corresponding to the engine load and the engine speed read in step S4 from the above map. In addition, the ECU 70 corrects the vertical turning weighting coefficient α in the map of the stored elapsed time and the vertical turning weighting coefficient α with the above correction coefficient, and corrects the vertical turning weighting coefficient α at each time point after the end of the main injection Jm. Calculate α.

After calculating the vertical turning weighting coefficient α at each time point after the end of the main injection Jm in this way, the ECU 70 substitutes this into equation (6). As a result, a waveform of oxygen concentration fluctuation at the swirling reference point Z as shown by line L1 in FIG. 11 is defined, which takes into account the effects of both the vertical swirling flow and the horizontal swirling flow.

Here, line L2 shown by a broken line in the graph of FIG. 11 is a horizontal oxygen concentration waveform, and line L3 shown by a chain line is a vertical oxygen concentration waveform. As shown by the difference between these lines L2 and L3, the period of the horizontal oxygen concentration waveform, that is, the frequency of the horizontal swirling flow (lateral swirling frequency f2), is the period of the vertical oxygen concentration waveform, that is, the frequency of the vertical swirling flow. frequency (vertical rotation frequency f1). This is because the distance between injection axes d, which is a parameter related to the frequency of the swirling flow in the horizontal direction, is shorter than the swirling distance D, which is a parameter related to the frequency of the swirling flow in the longitudinal direction. In particular, in this embodiment, the injector 15 has 10 nozzle holes 152, and the distance d between the injection axes is sufficiently small, so the horizontal swirl frequency f2 has a large value of more than twice the vertical swirl frequency f1. Become. In addition, when the elapsed time after the main injection Jm ends is short as described above, the vertical swirl weighting coefficient α is large, and the influence of the vertical swirl flow on the oxygen concentration at the swirl reference point Z is greater. . As a result, as shown by the solid line in FIG. 11, the oxygen concentration fluctuation waveform at the turning reference point Z has a shape that is relatively close to the vertical oxygen concentration waveform shown by the chain line.

<How to calculate injection interval time>
Next, a method for calculating the injection interval time Ti in step S6 will be described in detail. As mentioned above, in order to obtain the injection interval time Ti, it is necessary to determine the time when the middle part of the clean air flow E shown in FIG. It is necessary to specify the timing of a certain oxygen arrival. This oxygen arrival time is the time when the function x3(t) shown in the above equation (6) reaches its maximum value, and is the time when the oxygen concentration at the turning reference point Z takes the maximum value for the first time after the end of the main injection Jm. corresponds to

Here, the oxygen arrival timing according to the above equation (6) appears periodically (repeatedly) in calculation. On the other hand, from the viewpoint of fuel efficiency, it is preferable that the timing of after-injection Ja be as early as possible.

In step S6, the injection interval time Ti is calculated by a predetermined calculation using the above knowledge. That is, in the above step S6, the ECU 70 finds the earliest time among the times when the function x3(t) shown in the above equation (6) reaches the maximum value, and sets this as the oxygen arrival time. In the example of FIG. 11, time t11 corresponds to the oxygen arrival time. Then, the ECU 70 determines the injection interval time Ti so that the fuel spray Fa caused by the after-injection Ja reaches the turning reference point Z at the calculated oxygen arrival time (for example, time t11 in FIG. 11).

In order for the fuel spray Fa to reach the turning reference point Z at the oxygen arrival time, it is necessary to start the after injection Ja a little before the oxygen arrival time. That is, the time required for the fuel spray Fa caused by the after-injection Ja to move from the nozzle hole 152 to the turning reference point Z, that is, the time required from the start of the after-injection Ja until the tip of the spray Fa reaches the turning reference point Z. If time is defined as the required spray arrival time, it is necessary to set the after injection Ja start time to be earlier than the oxygen arrival time by the spray arrival required time. Therefore, the ECU 70 calculates the oxygen arrival timing calculated as described above, in other words, the time required from the end of the main injection Jm until the oxygen concentration at the turning reference point Z becomes the highest (the solid line waveform in FIG. 11). In this case, the value obtained by subtracting the above-mentioned time required to reach the spray from t11 (msec) is calculated as the injection interval time Ti. Note that the time required for the spray to reach the spray (the time required for the fuel spray Fa caused by the after-injection Ja to move from the nozzle hole 152 to the turning reference point Z) can be calculated each time, but it may be determined by a predetermined fixed value. may also be used. This is because the time required for the spray to reach the spray is relatively short, and it is thought that fluctuations due to differences in conditions are also small.

FIG. 19 is a group of graphs showing the relationship between the injection interval time Ti calculated as described above and each parameter of main injection amount, injection pressure, intake pressure, engine water temperature, and fuel temperature. As shown in this figure, the injection interval time Ti becomes shorter as the injection amount of main injection Jm increases, becomes shorter as the fuel injection pressure becomes higher, becomes longer as the intake pressure becomes higher, and becomes shorter as the engine water temperature becomes higher. , the higher the fuel temperature, the shorter the time. It is assumed that each graph shown in FIG. 19 shows a change in the injection interval time Ti obtained when the parameter shown on the horizontal axis changes alone (when the other parameters are constant). Further, each graph is a linear graph representing a simple direct proportional or inverse proportional relationship, but it is only a schematic and is not necessarily a linear graph.

Here, as described above, the lateral swirl frequency f2 and the lateral oxygen concentration waveform (function x2(t)) change only depending on the engine speed, and do not change depending on the main injection amount, injection pressure, intake pressure, and fuel temperature. From this, changes in the injection interval time Ti due to changes in each parameter shown in each graph of FIG. 19 are due to changes in the longitudinal oxygen concentration waveform (function x1(t)) due to changes in each parameter.

When only the longitudinal oxygen concentration waveform (function x1(t)) is considered, the time tx at which the oxygen concentration at the turning reference point Z becomes the highest can be expressed by the following equation (7).

t=(φ+nπ)/2πf1 (N=0, 2, 4,...) (7)
In other words, when there is no horizontal swirl flow, the oxygen arrival timing is when t, which is the elapsed time from the end of main injection Jm, satisfies the relationship of equation (7), and the longitudinal swirl frequency f1 and initial phase φ It can be expressed as a function with only variables. According to this equation (7), when there is no horizontal swirling flow, the timing of oxygen arrival becomes earlier as the vertical swirling frequency f1 becomes larger, and becomes later as the initial phase φ becomes larger.

Here, from equation (2), the longer the spray length L is, the larger the initial phase φ becomes. Based on the above findings, it is thought that the timing of oxygen arrival will be delayed. However, as the spray length L increases, the length of the clean air E in the swirling direction that occupies the swirling flow becomes shorter, and as a result, the period during which the clean air E exists at the swirling reference point Z becomes shorter, and as a result, the timing of oxygen arrival becomes earlier. . From this, regarding the engine water temperature and fuel temperature, as shown in the graph of FIG. 14(c), the spray length L becomes longer when these temperatures are higher, and the injection length L becomes longer as shown in the graph of FIG. 19. The interval time Ti is shortened.

<Engine speed and injection interval time>
The graph in FIG. 20 shows the relationship between the injection interval time Ti and the engine speed. The three lines in the same graph show the above relationships at different engine loads. Lines L11, L12, and L13 indicate that the engine load increases in this order, and line L13 indicates when the engine load reaches the maximum value in the diffusion combustion region A1 (when the engine load falls within the diffusion maximum load region A1_m). This is a graph of

As shown in FIG. 20, regardless of the engine load, the higher the engine speed, the shorter the injection interval time Ti. This is because the higher the engine speed, the earlier the oxygen concentration at the swirling reference point Z reaches its maximum value in both the longitudinal and lateral swirling flows.

Specifically, as shown in FIG. 14(b), as the engine speed increases, the turning distance D increases. As expressed by equation (2), as the turning distance D becomes longer, the initial phase φ of the longitudinal oxygen concentration waveform (function X1(t)) becomes smaller. From this, as the engine speed increases, the fluctuation waveform changes from the solid line in Figure 13 to the broken line in Figure 13, and the time when the oxygen concentration at the swirl reference point Z reaches its maximum value due to the vertical swirl flow becomes faster. In the example of FIG. 13, the time when the oxygen concentration reaches its maximum value is advanced from time t22 to time t122. Note that as the turning distance D increases, the vertical turning frequency f1 decreases, but the effect of the above-mentioned initial phase φ becoming smaller is greater than this, and as the engine speed increases, the above-mentioned oxygen concentration reaches its maximum. The time when it becomes a value will be earlier.

Furthermore, as shown in FIG. 16, as the engine speed increases, the speed VS of the swirl flow increases, and the horizontal rotation frequency f2 increases. As the engine speed increases, the fluctuation waveform changes from the solid line in FIG. 15 to the broken line in FIG. It's also faster. In the example of FIG. 15, the timing at which the oxygen concentration reaches its maximum value is advanced from time points t31 and t32 to time points t131 and t132. Therefore, the higher the engine speed, the earlier the oxygen concentration at the turning reference point Z reaches its maximum value, and the injection interval time Ti is shortened accordingly.

However, as shown by lines L11 and L12 in FIG. 20, in regions other than the diffusion maximum load region A1_m (FIG. 7), the injection interval time Ti for the increase in engine speed is longer in the high speed region A1_H than in the low speed region A1_L. The shortening rate of is increased. That is, in the remaining diffusion combustion region A1 excluding the diffusion maximum load region A1_m, the higher the engine speed, the greater the reduction time of the injection interval time Ti when the engine speed increases by the unit increase amount. This is because, as described above and as shown in FIG. 18, the vertical turning weighting coefficient α is smaller in the high speed region A1_H than in the low speed region A1_L.

Specifically, in the high-speed region A1_H, the vertical rotation weighting coefficient α is small, and the influence of the horizontal rotation flow on the oxygen concentration at the rotation reference point Z becomes large. As mentioned above, and as shown by the difference between the broken line L2 and the chain line L3 in FIG. 11, the swirling frequency in the horizontal direction is higher than that in the longitudinal direction, and The time when the oxygen concentration at the swirling reference point Z reaches its maximum due to the flow is earlier than the time when the oxygen concentration at the swirling reference point Z reaches its maximum due to the longitudinal swirling flow. Therefore, as the influence of the lateral swirling flow increases, that is, as the vertical swirling weighting coefficient α increases, the time point at which the oxygen concentration at the swirling reference point Z reaches its maximum approaches the point at which it reaches its maximum due to the lateral swirling flow. As a result, the time point at which the oxygen concentration at the turning reference point Z reaches its maximum becomes earlier.

Turning reference points Z at operating points C1, C2, C3 included in an area where the engine load is lower than the diffused maximum load area A1_m, and where the engine loads are the same and the engine speeds are different from each other. FIG. 21 shows a comparison of the oxygen concentration fluctuation waveforms. In FIG. 21, lines L21, L22, and L23 are waveforms of operating points C1, C2, and C3, respectively. The driving points C1 and C2 are driving points included in the low speed area A1_L, and the driving point C3 is a driving point included in the high speed area A1_H. Further, the difference in engine speed between operating points C1 and C2 is the same as the difference in engine speed between operating points C2 and C3. As shown in FIG. 21, the time when the oxygen concentration at the turning reference point Z reaches its maximum value is at time t41 at driving point C1, while at time t42 at driving point C2, which is earlier than time t41, and at the driving point At C3, the time point t43 is even earlier than the time point t42. The difference between the time t43 and the time t42 is longer than the difference between the time t42 and the time t41.

In this way, in the remaining diffusion combustion region A1 excluding the diffusion maximum load region A1_m, the vertical rotation weighting coefficient α is smaller when the engine speed is high than when it is low, so that the injection interval with respect to an increase in the engine speed is The reduction rate of time Ti is increased.

Further, as described above, in the high-speed region A1_H, the vertical turning weighting coefficient α is increased as the engine load is higher. From this, as shown by the difference in the slopes of lines L11, L12, and L13 in FIG. 20, in the high-speed region A1_H, the higher the engine load, the smaller the reduction rate of the injection interval time Ti with respect to the increase in engine speed. be done. Furthermore, in the diffusion maximum load region A1_m, since the vertical rotation weighting coefficient α is a constant value regardless of the engine speed, the reduction rate of the injection interval time Ti with respect to the increase in the engine speed is It is assumed to be constant.

<Engine water temperature and injection interval time>
As described above, and as shown in the graph of FIG. 19, the higher the engine water temperature is, the shorter the injection interval time Ti is. Therefore, as shown in the graph of FIG. 22, even if the engine speed is the same, the injection interval time Ti is shorter when the engine water temperature is high than when it is low. In the graph of FIG. 22, the solid line is the line when the engine water temperature is lower than the broken line.

<Fuel injection during acceleration operation>
Next, a specific example of the injection interval time Ti set during acceleration operation will be described. Here, as an example, we will explain the injection interval time Ti when the engine operating point changes from operating point C1 as shown by arrow B in the diffusion combustion region A1 to operating point C2 where the engine speed and engine load are higher than this. Explain the changes. Note that in the example of FIG. 5, the arrow B is just an example, and as long as the conditions are such that the injection amount of the main injection Jm increases during acceleration within the diffusion combustion region A1, the injection interval time Ti is as follows. Changes with similar trends.

FIG. 23 is a time chart showing temporal changes in various state quantities during acceleration operation within the diffusion combustion region A1. Time t51 in this figure is the time when the accelerator opening increases to a predetermined opening due to the driver's further depression of the accelerator pedal (see chart (a)). As the accelerator opening degree increases at time t51, the engine load (required torque) increases in a stepwise manner, and the injection amount of the main injection Jm and the target injection pressure of fuel increase in a stepwise manner accordingly. That is, the injection amount of main injection Jm changes in a stepwise manner from Q1 to Q2, which is larger than this (see chart (b)), and the target injection pressure of fuel changes in a stepwise manner from P1 to P2, which is larger than this. (See chart (c)). On the other hand, the actual injection pressure (actual injection pressure) gradually increases according to the operation of the fuel pressure regulator 16, and reaches P2, which is the same value as the target injection pressure after the increase, at time t52, which is later than time t51 ( (See chart (d)).

The increase in the injection amount of the main injection Jm after the above-mentioned time t51 increases the output torque of the engine and gradually increases the engine speed. In the illustrated example, the engine speed continues to increase until time t53, which is later than time t52 when the increase in actual injection pressure is completed, at which point the increase is completed. That is, the engine speed increases from N1 to N2, which is greater than this, from time t51 to time t52, and further increases from N2 to N3, which is greater than this, from time t52 to time t53 (see chart (e)). .

Further, since the increase in the output torque (heat generation amount) and the rotational speed increases the supercharging ability of the turbo supercharger 36, the intake pressure (supercharging pressure) gradually increases after time t51. In the illustrated example, the intake pressure continues to increase until a time t54, which is later than the time t53 when the increase in engine speed is completed, at which point the increase is completed. That is, the intake pressure increases from Ps1 to Ps3, which is greater than Ps1, from time t51 to time t53, and further increases from Ps3 to Ps4, which is greater than Ps3, from time t53 to time t54 (see chart (f)).

Due to the above-mentioned changes in injection amount, injection pressure, engine speed, and intake pressure over time, the injection interval time Ti, which is the time from the end of main injection Jm to the start of after injection Ja, is determined as shown in chart (g). Changes to That is, the injection interval time Ti decreases stepwise from T1 to T1' at time t51, gradually decreases from T1' to T2 from time t51 to time t52, and decreases from T2 to T3 from time t52 to time t53. and gradually increases from T3 to T4 from time t53 to time t54.

Among the above changes, the stepwise decrease in the injection interval time Ti at time t51 is due to the stepwise increase in the injection amount of main injection Jm shown in chart (b) from Q1 to Q2 at time t51. It is something. That is, as shown in FIG. 19, an increase in the injection amount of the main injection Jm has the effect of shortening the injection interval time Ti, so if the injection amount increases in a stepwise manner at time t51, the injection amount increases accordingly. The interval time Ti will be shortened in steps.

The gradual decrease in the injection interval time Ti from time t51 to time t52 is mainly due to the fact that the fuel injection pressure (actual injection pressure) shown in chart (d) gradually increases from P1 to P2 from time t51 to time t52. This is due to That is, as shown in FIG. 19, an increase in the injection pressure has the effect of shortening the injection interval time Ti, so if the injection pressure gradually increases from time t51 to t52, the injection interval time Ti will increase accordingly. will be gradually shortened.

The gradual decrease in the injection interval time Ti from time t52 to time t53 is mainly due to the engine speed shown in chart (e) gradually increasing from N2 to N3 from time t52 to time t53. That is, as shown in FIG. 19, an increase in the engine speed has the effect of shortening the injection interval time Ti, so if the engine speed gradually increases from time t52 to t53, the injection interval will change accordingly. The time Ti will be gradually shortened. Furthermore, as shown in FIG. 20, when the engine speed becomes higher than the reference speed N0, the reduction rate of the injection interval time Ti with respect to the increase in the engine speed increases (diffusion except for the diffusion maximum load area A1_m). In the combustion region A1), when the engine rotational speed exceeds the reference rotational speed N0 after a predetermined period of time has elapsed from time t52, the speed at which the injection interval time Ti is shortened increases.

The gradual increase in the injection interval time Ti from time t53 to time t54 is due to the fact that the intake pressure (supercharging pressure) shown in chart (f) gradually increases from Ps3 to Ps4 from time t53 to time t54. That is, as shown in FIG. 19, an increase in the intake pressure has the effect of lengthening the injection interval time Ti, so if the intake pressure gradually increases from time t53 to t54, the injection interval time Ti will increase accordingly. will be gradually extended.

<Effect>
As explained above, in this embodiment, the diffusion combustion of a diesel engine includes a piston 5 in which a reentrant cavity 5C is formed in the crown surface 50, and an injector 15 in which a plurality of nozzle holes 152 are formed in the tip part 151. When operating in region A1, main injection Jm injects the largest proportion of fuel into the first cavity part 51 out of the total injection amount during one combustion cycle, and a predetermined injection period during the expansion stroke that is delayed from main injection Jm. The injector 15 is controlled so that the after injection Ja, which injects a smaller amount of fuel than the main injection Jm, is executed at the same time, and the injection interval is the time from the end of the main injection Jm to the start of the after injection Ja. The time Ti is shortened as the engine speed increases, and the reduction rate of the injection interval time Ti as the engine speed increases is increased as the engine speed increases. According to such a configuration, it is possible to increase the air utilization rate of the fuel injected by the after-injection Ja and sufficiently suppress the generation of soot, while maintaining relatively good fuel efficiency.

That is, as shown in FIG. 10, the fuel spray Fm injected by the main injection Jm forms a vertical vortex along each wall surface of the lip portion 513, outer peripheral portion 512, and bottom portion 511 of the first cavity portion 51. The injector 15 then returns to the turning reference point Z on the injection axis AX (an extension of the central axis of the nozzle hole 152). The oxygen arrival time when the oxygen concentration at this turning reference point Z becomes high is the time when the clean air flow E (air flow with high oxygen content) that is generated following the rear end of the spray Fm passes through the turning reference point Z. If the fuel spray Fa caused by the after injection Ja can be made to reach the turning reference point Z in accordance with the oxygen arrival timing, the air utilization rate of the fuel injected by the after injection Ja can be increased (see FIG. 12). .

Here, the inventor's research has revealed that the timing of oxygen arrival becomes earlier as the engine speed increases, as shown in FIG. 20. As a control taking this point into consideration, in this embodiment, the injection interval time Ti, which is the time from the end of the main injection Jm to the start of the after injection Ja, is adjusted so that it becomes shorter as the engine speed increases. The fuel spray Fa caused by the after-injection Ja can be caused to reach the turning reference point Z at an appropriate time (that is, the time when the oxygen concentration at the turning reference point Z becomes high) according to the above-mentioned tendency of the oxygen arrival time, The air utilization rate of the fuel spray Fa can be increased. Furthermore, when the engine speed is high, the swirl flow speed VS increases and the horizontal swirling flow of the fuel spray becomes stronger, which has a greater effect on the timing of oxygen arrival, and this increases the engine speed. It is known that the timing of oxygen arrival becomes earlier with respect to increase. In contrast, in this embodiment, the injection interval time Ti is set such that the higher the engine speed is, the greater the reduction rate of the injection interval time Ti with respect to the increase in the engine speed. As a result, as shown by the solid line waveform in chart (h) of Fig. 23, compared to the case where the injection interval time Ti is set fixedly (shown by the two-dot chain line), the generation of soot due to combustion is reduced. can be effectively suppressed.

In addition, if the injection interval time Ti is variable according to the engine speed as described above, the injection timing of the after injection Ja can be advanced depending on the conditions compared to the case where the injection interval time Ti is fixed. It is possible to improve the fuel efficiency of the engine. For example, when the injection interval time Ti is set constant regardless of the engine speed, the temperature of the combustion chamber 6 is sufficient so that the amount of soot generated does not become excessive regardless of whether the engine speed is high or low. It is necessary to start the after-injection Ja after waiting for the fuel injection temperature to decrease to , that is, after a relatively long time has elapsed since the end of the main injection Jm (the expansion stroke has progressed to some extent). This reduces the proportion of the combustion energy based on after-injection Ja that is used as work, leading to deterioration in fuel efficiency. On the other hand, when the injection interval time Ti is made variable according to the engine speed as in the above embodiment, there is no need to uniformly delay the start time of the after injection Ja as described above, and the condition Depending on the situation, the injection timing of after injection Ja can be advanced. Thereby, the rate at which combustion energy based on after-injection Ja is converted into work can be increased as much as possible, and the fuel efficiency of the engine can be improved.

Further, in the above embodiment, the main injection Jm is executed at the timing when the injection axis AX of the injector 15 and the lip part 513 of the first cavity part 51 intersect, and after the end of this main injection Jm, the injection axis AX is The oxygen arrival time, which is the time when the clean air flow E reaches the turning reference point Z, is calculated based on multiple parameters (main injection amount, injection pressure, intake pressure, engine water temperature, fuel temperature) including engine speed. The injection interval time Ti is determined based on the calculated oxygen arrival timing. According to such a configuration, the oxygen arrival time can be appropriately calculated based on the knowledge obtained by the present inventor that the oxygen arrival time changes depending on a specific parameter group including the engine rotation speed, and the oxygen arrival time calculated By adjusting the start timing (injection interval time Ti) of after-injection Ja so that the fuel spray Fa caused by after-injection Ja reaches the turning reference point Z in accordance with the timing, the air utilization rate of the fuel spray Fa can be increased. The amount of soot generated can be reduced.

Furthermore, in the embodiment described above, the injection interval time Ti is adjusted in consideration of various parameters other than the engine speed. Specifically, the injection interval time Ti is adjusted to be shorter as the main injection amount, injection pressure, engine water temperature, or fuel temperature is higher, and adjusted to be longer as the intake pressure is higher. According to such a configuration, it is possible to set an appropriate injection interval time Ti that takes into account various parameters that change the timing of oxygen arrival, and to start after-injection Ja at an appropriate time when a high air utilization rate can be obtained. be able to.

In particular, if the engine water temperature is different, the timing of oxygen arrival will be different even if the engine load and rotation speed are the same. However, in the above embodiment, the injection amount, injection pressure, intake pressure, etc. Since the timing of oxygen arrival is determined by taking into account not only parameters depending on the rotation speed but also the engine water temperature , good air utilization can be achieved under various temperature conditions in the combustion chamber 6 that change depending on the degree of warming up. rate can be secured without any problem. As mentioned above, the difference in engine water temperature affects the timing of oxygen arrival because the higher the engine water temperature, the longer the length of the fuel spray (spray length L), and the longer the clean air flow E becomes. This is because the timing at which the intermediate portion of the clean air flow E passes the turning reference point Z is earlier due to the shorter length. Therefore, in the embodiment described above, the higher the engine water temperature is, the shorter the injection interval time Ti is, thereby improving the air utilization rate.

<Modified example>
In the above embodiment, an example in which the present invention is applied to a diesel engine equipped with a piston 5 in which an upper and lower two-stage cavity 5C including a first cavity part 51 and a second cavity part 52 is formed on the crown surface 50 will be described. However, a diesel engine to which the present invention is applicable may include a piston having a single-stage cavity instead of a two-stage cavity. That is, in the cavity 5C of the piston 5 of the above embodiment, a chevron-shaped bottom portion 511, an outer peripheral portion 512 that is concave so as to protrude radially outward, and a lip portion that protrudes radially inwardly. The present invention can be applied to diesel engines equipped with pistons of various shapes, as long as the piston has at least a reentrant cavity corresponding to the first cavity portion 51 having the first cavity portion 513.

In the above embodiment, each of the plurality of injectors 15 provided in each of the plurality of cylinders 2 is provided with the injection pressure sensor SN5, and when injecting fuel from the injector 15 in any cylinder 2, the injector for that cylinder 2 Although the injection interval time Ti is determined based on the injection pressure detected at the IVC time point (the closing timing of the intake valve) by the injection pressure sensor SN5 provided in the The detection method is not limited to this. For example, an injection pressure sensor may be provided on the common rail 18 connected to the plurality of injectors 15 via the fuel supply pipe 17, and the injection interval time Ti may be determined based on the injection pressure detected by the injection pressure sensor. In addition, the injection pressure used to determine the injection interval time Ti in a certain cylinder 2 is determined before the injector 15 for the cylinder 2 injects fuel and in the cylinder one combustion order before the cylinder 2. It may be done after the combustion in Step 2 is completed, and is not limited to the time of IVC.

In the above embodiment, the oxygen arrival time (for example, the solid line in FIG. In the case of the waveform L1, the injection interval time Ti is adjusted so that the fuel spray Fa caused by the after-injection Ja reaches the turning reference point Z in accordance with this oxygen arrival timing. The arrival time is the time when any of the main parts of the clean air flow E excluding the front end and the rear end passes the turning reference point Z (the time when the period in which the fuel spray Fm exists on the turning reference point Z is clearly avoided) ), and there is no need to limit it to the period when the oxygen concentration is the highest. In particular, during cold operation when the engine water temperature is sufficiently low, the stability of combustion based on after-injection Ja is likely to decrease, so the start timing of after-injection Ja should be brought forward as much as possible in order to ensure the combustion stability. This may be required. In such a case, it is preferable to adjust the injection interval time Ti so that the fuel spray Fa caused by the after-injection Ja reaches the turning reference point Z at a timing slightly earlier than the timing when the oxygen concentration becomes the highest.

In the above embodiment, the number of main injections Jm is set to one, and the main injection Jm is executed in such a manner that fuel is continuously injected over a predetermined period including the compression top dead center. , a relatively large amount of fuel may be injected at a timing such that the fuel spray heads toward the lip portion 513 of the cavity 5C (that is, near the compression top dead center), and the number of injections is not limited to one. . For example, the main injection may be performed at multiple timings near the compression top dead center.

The number of nozzle holes 152 of the injector 15 is not limited to the above.

2 cylinder 5 piston 5C cavity 6 combustion chamber 15 injector 50 crown (of piston) 70 ECU (injection control unit)
152 (injector) nozzle hole 511 bottom 512 outer periphery 513 lip SN5 injection pressure sensor A1 diffuse combustion region (predetermined operating region)
AX Injection axis C1 1st operating point C2 2nd operating point E Clean air flow Fm Fuel spray (by main injection) Fa Fuel spray (by after injection) Jm Main injection Ja After injection Q1 1st injection amount Q2 2nd injection amount P1 First injection pressure P2 Second injection pressure Ti Injection interval time Z Turning reference point (specific position on the injection axis)

Claims (2)

A device for controlling a diesel engine including a cylinder, a piston housed in the cylinder so as to be able to reciprocate, and an injector that injects fuel containing light oil into a combustion chamber that is a space above the piston,
an injection control unit that controls the injector so that fuel is injected at a plurality of timings set from a compression stroke to an expansion stroke when operating in a predetermined operating region;
The piston has a cavity recessed downward in the crown surface thereof, and a bottom portion formed such that the height becomes lower toward the radially outer side as a wall surface defining the cavity, and a bottom portion formed radially outwardly than the bottom portion. A curved outer peripheral part that is curved and concave so as to be convex radially outward in a cross-sectional view including the cylinder axis, and a curved outer peripheral part that is formed above the outer peripheral part and convex radially inward in a cross-sectional view including the cylinder axis. It has a curved lip part that protrudes from the
The injector has a plurality of nozzle holes in the ceiling of the combustion chamber at a position facing the center of the cavity, and injects fuel diagonally downward radially and radially outward from each of the nozzle holes. It is set up so that
During operation in the predetermined operating range, the injection control section injects the largest proportion of fuel out of the total amount of fuel injected during one combustion cycle, and directs the injected fuel toward the lip section. A main injection that redirects at least a portion of the fuel downward from the lip portion, and an after injection that injects a smaller amount of fuel than the main injection at a predetermined time during an expansion stroke that is later than the main injection, to the injector. let it run,
The injection interval time, which is the time from the end of the main injection to the start of the after injection, is shortened as the engine speed increases, and the reduction rate of the injection interval time with respect to the increase in the engine speed is as follows: 1. A control device for a diesel engine, characterized in that the higher the engine rotation speed is, the higher the rotation speed is, except when the speed corresponds to a maximum value within a predetermined operating range . The diesel engine control device according to claim 1,
further comprising a temperature sensor that detects a predetermined temperature parameter that increases as the engine warms up ;
The control device for a diesel engine, wherein the injection control unit shortens the injection interval time as the temperature parameter detected by the temperature sensor increases, under the condition that parameters other than the temperature parameter are the same. .