JP2012172664A - Control device for spark ignition type gasoline engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To smooth the transition of modes while avoiding a problem occurring in a mode transition period, in a spark ignition type gasoline engine which performs the switching of modes between a compression ignition mode for executing compression ignition combustion and a spark ignition mode for executing spark ignition combustion.SOLUTION: The control device (PCM 10) selects a compression ignition mode in a low load region, and selects a spark ignition mode in a high load region which relatively increases fuel pressure and drives a fuel ejection valve 67 so as to include fuel ejection executed at predetermined timing during a retard period from a later period of a compression stroke to an initial period of an expansion stroke, and then ignites fuel after the ejection. The control device also selects a switching mode which ejects fuel at fuel pressure in the spark ignition mode and at delayed timing later than predetermined timing during the predetermined transition period in switching the mode between the compression ignition mode and the spark ignition mode accompanied by a change of a load, and ignites the fuel after the ejection.

Description

  The technology disclosed herein relates to a control device for a spark ignition gasoline engine.

  For example, as described in Patent Document 1, an HCCI engine that compresses and ignites a lean air-fuel mixture is known as a technique for achieving both improvement in exhaust emission and improvement in thermal efficiency. However, as the engine load increases, the compression ignition combustion becomes the combustion of premature ignition with a strong pressure rise. As a result, combustion noise increases and abnormal combustion such as knocking occurs, and Raw NOx increases due to a high combustion temperature. Therefore, in such an HCCI engine, compression ignition combustion is limited only to an operation region on the low load side, and in the operation region on the high load side, spark ignition combustion is generally performed by driving a spark plug.

JP 2009-91994 A

  By the way, the compression ignition combustion and the spark ignition combustion are greatly different in combustion form, and accordingly, various parameters relating to the operation of the engine such as the intake charge amount and the air-fuel ratio may be greatly different. Therefore, when changing the combustion mode from compression ignition combustion to spark ignition combustion in accordance with a change in engine load, and conversely, when changing the combustion mode from spark ignition combustion to compression ignition combustion, for example, an intake valve The operation of at least one of various valves such as the exhaust valve, the throttle valve, and the EGR valve is greatly changed.

  However, the control responsiveness related to the operation change of these various valves is greatly delayed with respect to the switching of the combustion mode. For this reason, various inconveniences may occur when switching the combustion mode. For example, Patent Document 1 points out that although intake throttling is performed when switching from compression ignition combustion to spark ignition combustion, knocking is likely to occur due to a response delay. Therefore, in Patent Document 1, it is described that, in order to avoid knocking, control combining EGR and fuel enrichment is executed during the delay.

  The technology disclosed herein has been made in view of such a point, and an object thereof is a mode between a compression ignition mode in which compression ignition combustion is performed and a spark ignition mode in which spark ignition combustion is performed. In a spark-ignition gasoline engine that performs switching, a problem that may occur during a mode transition period associated with engine load fluctuations is avoided, and mode transition is made smooth.

  A spark ignition type gasoline engine control device disclosed herein includes an engine body having a cylinder having a geometric compression ratio set to 14 or more and configured to be supplied with fuel containing at least gasoline. A fuel injection valve configured to inject the fuel into the cylinder, a spark plug disposed to face the cylinder and configured to ignite an air-fuel mixture in the cylinder, A fuel pressure variable mechanism configured to change a pressure of fuel injected by the fuel injection valve, and at least the fuel injection valve, the spark plug, and the fuel pressure variable mechanism are controlled to operate the engine body. And a controller configured as described above.

And the controller is
When the operating state of the engine body is in a predetermined low load region, a compression ignition mode for executing compression ignition combustion is set,
In the high load region where the load is higher than that in the compression ignition mode, the fuel pressure variable mechanism is controlled so that the fuel pressure is higher than at least in the low load region in the compression ignition mode, and the expansion stroke is started from the late stage of the compression stroke. Spark ignition that drives the fuel injection valve so as to include at least a fuel injection performed at a specific timing within the retard period until and drives the spark plug so as to ignite after the fuel is injected within the retard period Mode.

  The controller also includes a fuel pressure in the spark ignition mode within a predetermined transition period when switching between the compression ignition mode and the spark ignition mode in accordance with a change in the load of the engine body, and The switching mode is such that the fuel injection valve is driven so as to inject fuel at a timing retarded from the specific timing, and the spark plug is driven after the fuel injection.

  Here, the geometric compression ratio of the engine body may be set to 14 or more and, for example, 20 or less.

  In addition, the “low load range” and the “high load range” are respectively a low load side region and a high load side region when the operating region of the engine main body is divided into two regions for high and low loads. Good.

  The “late compression stroke” may be the late phase when the compression stroke is divided into three periods, an initial phase, a middle phase, and a late phase. Similarly, the “expansion stroke initial phase” refers to the expansion stroke as the initial phase, the middle phase, and It is good also as the initial stage when it divides into three periods of the latter term.

  First, when the operating state of the engine body is in a predetermined low load region, the compression ignition mode is executed in which compression ignition combustion is executed. Compression ignition combustion is advantageous for improving both exhaust emission and thermal efficiency. In particular, this engine body is advantageous in stabilizing compression ignition combustion because the compression end temperature and the compression end pressure in the cylinder are increased by setting the geometric compression ratio to 14 or more.

  On the other hand, when the operating region of the engine body is in a higher load region than the compression ignition mode, the spark ignition mode is set. In the high load range, the pressure and temperature in the cylinder are high, and abnormal combustion such as pre-ignition and knocking is likely to occur. Further, as described above, since the engine body has a high compression ratio, it is disadvantageous for occurrence of abnormal combustion such as pre-ignition and knocking.

  Therefore, in the engine control apparatus having the above-described configuration, in the spark ignition mode, the fuel pressure is increased, and fuel is injected during the retard period from the late stage of the compression stroke to the early stage of the expansion stroke. Avoid burning. This is due to the following reason.

  That is, the high fuel pressure relatively increases the amount of fuel injected per unit time. For this reason, when compared with the same fuel injection amount, the high fuel pressure shortens the period during which fuel is injected into the cylinder, that is, the injection period, as compared with the low fuel pressure.

  Further, the high fuel pressure is advantageous for atomization of the fuel spray injected into the cylinder and makes the flight distance of the fuel spray longer. For this reason, the high fuel pressure shortens the period (fuel mixture formation period) until the combustible air-fuel mixture is formed around the spark plug after the fuel injection is completed.

  Since the end timing of the mixture formation period is substantially the same as the ignition timing set near the compression top dead center, the shortening of the injection period and the shortening of the mixture formation period are the fuel injection timing (from Precisely, the injection start timing) can be set to a relatively late timing. That is, fuel injection is possible at a timing within the retard period from the latter stage of the compression stroke to the early stage of the expansion stroke.

  Further, as fuel is injected into a cylinder at a high fuel pressure, the turbulence in the cylinder becomes stronger and the turbulence energy in the cylinder increases. This high turbulent energy contributes to shortening of the combustion period, coupled with the fact that the fuel injection timing is set to a relatively late timing.

  In other words, even if fuel is injected into the cylinder at a high fuel pressure, if the injection timing is during the intake stroke, the time until the ignition timing is long or the inside of the cylinder is compressed during the compression stroke after the intake stroke. Therefore, the turbulence is attenuated, and the turbulent energy in the cylinder during the combustion period becomes relatively low. Higher turbulence energy in the cylinder is advantageous for shortening the combustion period, so even if fuel is injected into the cylinder at a high fuel pressure, the combustion period is shortened as long as the injection timing is during the intake stroke. Does not contribute greatly.

  On the other hand, injecting fuel into the cylinder at a relatively late timing within the retard period and at a high fuel pressure as in the above configuration starts combustion while suppressing attenuation of turbulence in the cylinder. This makes it possible to increase the turbulence energy in the cylinder during the combustion period. This shortens the combustion period.

  As described above, performing fuel injection into the cylinder at a high fuel pressure and a retard period with a relatively late timing reduces the injection period, the mixture formation period, and the combustion period. enable.

  Here, the sum of the injection period, the mixture formation period, and the combustion period corresponds to the reaction possible time of the unburned mixture. The longer this time is, the more the reaction of the unburned mixture proceeds. It tends to cause premature ignition and knocking. On the other hand, as described above, shortening the injection period, the mixture formation period, and the combustion period respectively significantly shortens the reaction time of the unburned mixture, so that abnormalities such as pre-ignition and knocking occur. Combustion can be effectively avoided.

  Since this configuration avoids abnormal combustion by devising the form of fuel injection into the cylinder, there is no need to retard the ignition timing for the purpose of avoiding abnormal combustion, or the amount of retardation can be reduced. It becomes possible to make it smaller. Since this makes it possible to advance the ignition timing as much as possible, in the spark ignition mode, the above configuration avoids abnormal combustion, improves thermal efficiency and torque, in other words, improves fuel efficiency. To be advantageous.

  As described above, in the engine control device disclosed herein, the combustion mode is greatly different between the compression ignition mode and the spark ignition mode, including the difference in fuel pressure. For this reason, when the engine load is changed, when switching from the compression ignition mode to the spark ignition mode and when switching from the spark ignition mode to the compression ignition mode, for example, execution / non-execution of internal EGR, execution of external EGR -Change control regarding various filling amounts, such as non-execution and intake throttle execution / non-execution, may be required. This creates a need to greatly change at least one operation of various valves such as an intake valve, an exhaust valve, a throttle valve, and an EGR valve. However, the control responsiveness of these operation changes is greatly delayed with respect to mode switching. For this reason, for example, when the fuel injection amount is set so as to satisfy the air-fuel ratio λ = 1 within the transition period when the mode is switched, the change control relating to the filling amount is not completed, so that the fresh air The amount may increase temporarily, and the fuel injection amount increases accordingly, and the torque may increase.

  Therefore, in the above-described configuration, the switching mode is set in which fuel is injected at a timing delayed from the specific timing set in the spark ignition mode within the transition period, and ignition is performed after the fuel injection. In the switching mode, the generated torque is suppressed because the ignition timing is largely retarded. On the other hand, it is usually disadvantageous in terms of combustion stability that the fuel injection timing and the ignition timing are greatly delayed, but the above-described fuel injection at a high fuel pressure increases the turbulent energy in the cylinder. This shortens the combustion period, so that the combustion in the switching mode is stabilized. As a result, it is possible to smoothly change the mode while avoiding a torque shock when the mode is switched between the compression ignition mode and the spark ignition mode. Then, when the change control related to the filling amount is completed, the mode is switched from the switching mode to the spark ignition mode or the compression ignition mode. Therefore, the “transition period” here can be defined as a period (time) corresponding to various control delays.

  The controller may operate the engine body at an air-fuel ratio λ = 1 in the spark ignition mode, and operate the engine body at a leaner air-fuel ratio λ = 1 in the compression ignition mode. Good.

  The operation at the stoichiometric air-fuel ratio in the spark ignition mode shortens the combustion period compared with the lean operation. Therefore, in the spark ignition mode, the reaction possible time of the unburned mixture is shortened, which is advantageous for avoiding abnormal combustion. In addition, operating the engine body at the stoichiometric air-fuel ratio makes it possible to use a three-way catalyst. For this reason, the operation of λ = 1 is advantageous for improving the emission performance in the spark ignition mode. On the other hand, operation at a lean air-fuel ratio in the compression ignition mode is advantageous for reducing NOx.

  Since the air-fuel ratio is different between the compression ignition mode and the spark ignition mode, it is necessary to change the intake charge amount during the transition period. In this case, the switching mode described above smoothly switches the mode between the compression ignition mode and the spark ignition mode while avoiding torque shock.

  The controller may increase the fuel injection amount from a fuel injection amount corresponding to an engine load in the switching mode.

  That is, in the switching mode, increasing the fuel injection amount beyond the fuel injection amount corresponding to the load causes an unnecessary increase in the torque. Suppresses torque rise. As a result, torque shock when the mode is switched as the engine load changes is suppressed.

  In the compression ignition mode, the controller performs internal EGR control in which a part of burned gas exists in the cylinder, and in the switching mode, the air-fuel ratio is λ == without performing the internal EGR control. The air-fuel ratio control may be performed so as to be 1.

  Executing the internal EGR control in the compression ignition mode effectively increases the compression end temperature in the cylinder and is advantageous for stabilizing the compression ignition combustion. When switching from the compression ignition mode to the spark ignition mode via the switching mode, when switching from the compression ignition mode to the switching mode, the internal EGR control is stopped to stop the internal EGR that has been introduced into the cylinder until then. Instead of gas, a lot of fresh air will be introduced into the cylinder (temporarily). On the other hand, in order to satisfy the theoretical air-fuel ratio λ = 1 in the switching mode, a large amount of fuel must be injected to meet the large amount of fresh air. As a result, the fuel injection amount temporarily increases in the switching mode, and the torque increases rapidly. However, in the switching mode, as described above, torque generation is suppressed by retarding the fuel injection timing and the ignition timing, so that it is possible to avoid a torque shock at the time of mode switching. Conversely, when shifting from the spark ignition mode to the compression ignition mode via the switching mode, in preparation for starting the internal EGR control in the compression ignition mode, the intake charge amount (fresh air amount) is gradually increased in the switching mode. Must. Even in this case, in order to satisfy the theoretical air-fuel ratio λ = 1 in the switching mode, a large amount of fuel must be injected to meet the large amount of fresh air, and the fuel injection amount increases and torque shock occurs. Therefore, as described above, it is possible to avoid a torque shock at the time of mode switching by retarding the fuel injection timing and the ignition timing.

  The fuel pressure variable mechanism may set the fuel pressure to 40 MPa or more in the spark ignition mode and the switching mode.

  A fuel pressure of 40 MPa or more effectively realizes all of the shortening of the injection period, the shortening of the mixture formation period, and the shortening of the combustion period as described above. As a result, it is possible to avoid abnormal combustion in the spark ignition mode and to stabilize combustion in the switching mode. In addition, what is necessary is just to set the maximum value of a fuel pressure according to the property of a fuel. As an example, but not limited to this, the maximum value of the fuel pressure may be set to about 120 MPa.

  The fuel injection valve may have a plurality of nozzle holes, and may be configured such that fuel spray injected from these nozzle holes spreads radially in the cylinder.

  A fuel injection valve having a plurality of injection holes is advantageous for improving the turbulent energy in the cylinder. Therefore, the fuel injection valve having a plurality of injection holes is effective for shortening the combustion period in the spark ignition mode and the switching mode, and thus stabilizing the combustion.

  In the spark ignition mode, the controller drives the fuel injection valve to execute a plurality of fuel injections within the retard period, and in the switching mode, the controller The timing of the last fuel injection may be delayed.

  In the spark ignition mode, fuel injection that is executed at a relatively early timing among a plurality of fuel injections that are executed in a divided manner can ensure a long mixture formation period. To be advantageous. Since a sufficient mixture formation period is ensured in such a manner, fuel injection that is executed at a relatively late timing can be executed at a more retarded timing. This is advantageous for improving the turbulent energy in the cylinder, and the combustion period is further shortened. That is, it is advantageous for avoiding abnormal combustion. Similarly, in the switching mode, the fuel injection executed at a relatively early timing among the plurality of fuel injections executed in a divided manner is advantageous for the vaporization atomization of the fuel. By retarding the timing of the last fuel injection among the injections, it is advantageous for suppressing the torque while stabilizing the combustion.

  In the compression ignition mode, the controller drives the fuel injection valve so as to inject fuel at a timing more advanced than the fuel injection timing in the retard period in the spark ignition mode, and The air-fuel mixture in the cylinder may be compressed and ignited by causing a part of the burned gas of the engine body to exist in the cylinder.

  In the compression ignition mode, a relatively homogeneous lean air-fuel mixture can be formed by advancing the fuel injection timing, and a part of burnt gas can be present in the cylinder to cause the compression end. Since the temperature increases, compression ignition combustion is stabilized. This is advantageous for improving fuel consumption.

  As described above, the control device for the spark ignition type gasoline engine switches between the compression ignition mode and the spark ignition mode according to the engine load, and in the spark ignition mode, the compression is performed with a high fuel pressure. Abnormal combustion can be effectively avoided by performing fuel injection and performing spark ignition combustion at a timing within the retard period from the latter stage of the stroke to the early stage of the expansion stroke.

  In the transition period of mode switching accompanying changes in the engine body load, the fuel injection timing and ignition timing are set to a switching mode that retards more than in the spark ignition mode, thereby suppressing an increase in torque. Torque shock is prevented.

It is the schematic which shows the structure of a spark ignition type gasoline engine. It is a block diagram concerning control of a spark ignition type gasoline engine. It is sectional drawing which expands and shows a combustion chamber. It is a figure which illustrates the operating area of an engine. It is a figure which compares the state of SI combustion by a high pressure retarded injection, and the state of conventional SI combustion. Figure showing the difference between SI combustion by high pressure retarded injection and conventional SI combustion (upper figure) in the relationship between the unburned mixture reaction possible time and the unburned mixture reaction progress at the end of combustion; It is a figure (respective figure of a middle stage and each lower stage) which shows the relationship between a fuel pressure and each parameter related to unburned air-fuel mixture reaction possible time. It is a figure which shows the difference between SI combustion by high pressure retarded injection, and conventional SI combustion in the relationship between an ignition timing and the unburned air-fuel mixture reaction progress at the combustion end timing. It is a figure which shows the difference between SI combustion by high pressure retarded injection, and the conventional SI combustion in the relationship between ignition timing, thermal efficiency, and torque. It is a figure which shows the difference of (a) heat generation rate (dQ / d (theta)) and (b) in-cylinder pressure rise rate (dP / d (theta)) of SI combustion and CI combustion by high pressure retarded injection. It is a timing chart which shows the difference in operation | movement of an intake valve and an exhaust valve with respect to the difference in an engine load, and the difference in an ignition timing and an injection timing. (A) Mixture filling amount, (b) Throttle valve opening, (c) when controlling the internal EGR amount by controlling the intake valve in the low load region and performing throttle control of the intake air in the high load region Changes in EGR valve opening, (d) exhaust valve closing timing, (e) intake valve opening timing, (f) intake valve closing timing, (g) intake valve lift amount It is a figure which shows an example, respectively. FIG. 12 is a diagram corresponding to FIG. 11 when the internal EGR amount is controlled by controlling the throttle valve in the low load range and the intake throttle control is performed in the high load range. FIG. 12 is a diagram corresponding to FIG. 11 when the internal EGR amount is controlled by controlling the intake valve in the low load region and the external EGR is used in the high load region. FIG. 12 is a diagram corresponding to FIG. 11 when the internal EGR amount is controlled by controlling the throttle valve in the low load range and the external EGR is used in the high load range. (A) Mixture filling amount, (b) G / F, (c) Injection timing, (d) Fuel when the internal EGR amount is controlled in the low load region and the intake air is throttled in the high load region It is a figure which shows an example of a change of a pressure, (e) injection pulse width, and (f) ignition timing, respectively. FIG. 16 is a diagram corresponding to FIG. 15 in a case where the internal EGR amount is controlled in the low load range and the external EGR is used in the high load range. It is a figure which illustrates the mode change accompanying the change of an engine load, and the transitional change of the structure of the air-fuel mixture in a cylinder. It is a flowchart of the engine control which PCM performs. It is a flowchart which concerns on control of CI mode included in the flow of FIG. It is a flowchart which concerns on control of SI mode contained in the flow of FIG. FIG. 21 is a characteristic diagram of parameters calculated in steps included in the flows of FIGS. 18 to 20. It is a flowchart which concerns on control of the switching mode contained in the flow of FIG. It is a characteristic view of the parameter calculated in the step included in the flow of FIG.

  Hereinafter, an embodiment of a control device for a spark ignition gasoline engine will be described with reference to the drawings. The following description of preferred embodiments is exemplary. 1 and 2 show a schematic configuration of an engine (engine body) 1. The engine 1 is a spark ignition gasoline engine that is mounted on a vehicle and supplied with fuel containing at least gasoline. The engine 1 is provided with a cylinder block 11 provided with a plurality of cylinders 18 (only one shown), a cylinder head 12 provided on the cylinder block 11, and a cylinder block 11 below the cylinder block 11. And an oil pan 13 in which oil is stored. A piston 14 connected to the crankshaft 15 via a connecting rod 142 is fitted in each cylinder 18 so as to be able to reciprocate. A cavity 141 having a reentrant shape is formed on the top surface of the piston 14 as shown in an enlarged view in FIG. The cavity 141 faces a direct injection injector 67 described later when the piston 14 is positioned near the compression top dead center. The cylinder head 12, the cylinder 18, and the piston 14 having the cavity 141 define a combustion chamber 19. The shape of the combustion chamber 19 is not limited to the illustrated shape. For example, the shape of the cavity 141, the top surface shape of the piston 14, the shape of the ceiling portion of the combustion chamber 19, and the like can be changed as appropriate.

  The engine 1 is set to a relatively high geometric compression ratio of 14 or more for the purpose of improving the theoretical thermal efficiency, stabilizing the compression ignition combustion described later, and the like. In addition, what is necessary is just to set a geometric compression ratio suitably in the range of about 14-20.

  The cylinder head 12 is provided with an intake port 16 and an exhaust port 17 for each cylinder 18. The intake port 16 and the exhaust port 17 have an intake valve 21 and an exhaust for opening and closing the opening on the combustion chamber 19 side. Each valve 22 is disposed.

  Among the valve systems that drive the intake valve 21 and the exhaust valve 22, respectively, on the exhaust side, the operation mode of the exhaust valve 22 is switched between a normal mode and a special mode, for example, a hydraulically operated variable mechanism (see FIG. 2). Hereinafter, a VVL (Variable Valve Lift) 71 is provided. Although the VVL 71 is not shown in detail in its configuration, two types of cams having different cam profiles, a first cam having one cam crest and a second cam having two cam crests, and its first And a lost motion mechanism that selectively transmits an operating state of one of the second cams to the exhaust valve. When the operating state of the first cam is transmitted to the exhaust valve 22, the exhaust valve 22 operates in the normal mode in which the valve is opened only once during the exhaust stroke (see FIGS. 10C and 10D). On the other hand, when the operating state of the second cam is transmitted to the exhaust valve 22, the exhaust valve 22 is opened during the exhaust stroke and is also opened during the intake stroke, so-called exhaust double opening. The operation is performed in a special mode (see FIGS. 10A and 10B). The normal mode and the special mode of the VVL 71 are switched according to the operating state of the engine. Specifically, the special mode is used in the control related to the internal EGR. In order to enable switching between the normal mode and the special mode, an electromagnetically driven valve system that drives the exhaust valve 22 by an electromagnetic actuator may be employed. Also, the execution of internal EGR is not realized only by opening the exhaust twice. For example, the internal EGR control may be performed by opening the intake valve 21 twice or by opening the intake valve twice, or by providing a negative overlap period in which both the intake valve 21 and the exhaust valve 22 are closed in the exhaust stroke or the intake stroke. Internal EGR control that causes the fuel gas to remain in the cylinder 18 may be performed.

  As shown in FIG. 2, the exhaust side valve system having the VVL 71 is arranged on the intake side as shown in FIG. 2. 72) and a lift variable mechanism (hereinafter referred to as CVVL (Continuously Variable Valve Lift)) 73 capable of continuously changing the lift amount of the intake valve 21. . The VVT 72 may employ a hydraulic, electromagnetic, or mechanical structure as appropriate, and illustration of the detailed structure is omitted. The CVVL 73 can also adopt various known structures as appropriate, and the detailed structure is not shown. With the VVT 72 and the CVVL 73, the intake valve 21 can change its valve opening timing, valve closing timing, and lift amount, as shown in FIGS.

  Further, for each cylinder 18, a direct injection injector 67 that directly injects fuel into the cylinder 18 and a port injector 68 that injects fuel into the intake port 16 are attached to the cylinder head 12.

  As shown in an enlarged view in FIG. 3, the direct injection injector 67 is disposed so that its injection hole faces the inside of the combustion chamber 19 from the central portion of the ceiling surface of the combustion chamber 19. The direct injection injector 67 directly injects an amount of fuel into the combustion chamber 19 at an injection timing according to the operating state of the engine 1 and according to the operating state of the engine 1. In this example, the direct injection injector 67 is a multi-injector type injector having a plurality of injection holes, although detailed illustration is omitted. Thereby, the direct injection injector 67 injects the fuel so that the fuel spray spreads radially. As indicated by the arrows in FIG. 3, the fuel spray injected radially from the central portion of the combustion chamber 19 at the timing when the piston 14 is located near the compression top dead center is a cavity formed on the top surface of the piston. By flowing along the wall surface of 141, it reaches around the spark plug 25 described later. It can be paraphrased that the cavity 141 is formed so that the fuel spray injected at the timing when the piston 14 is located near the compression top dead center is contained therein. The combination of the multi-hole injector 67 and the cavity 141 is advantageous in that it shortens the time until the fuel spray reaches around the spark plug 25 after fuel injection and shortens the combustion period. is there. In addition, the direct injection injector 67 is not limited to a multi-injection type injector, and an external valve-opening type injector may be adopted as the direct injection injector.

  As shown in FIG. 1, the port injector 68 is arranged facing an independent passage communicating with the intake port 16 to the intake port 16 and injects fuel into the intake port 16. One port injector 68 may be provided for each cylinder 18, and if two intake ports 16 are provided for one cylinder 18, the port injector 68 is provided for each of the two intake ports 16. Also good. The type of the port injector 68 is not limited to a specific type, and various types of injectors can be appropriately employed.

  The fuel tank (not shown) and the direct injection injector 67 are connected to each other by a high-pressure fuel supply path. A high-pressure fuel supply system 62 that includes a high-pressure fuel pump 63 and a common rail 64 and supplies fuel to the direct injection injector 67 at a relatively high fuel pressure is interposed on the high-pressure fuel supply path. The high-pressure fuel pump 63 pumps fuel from the fuel tank to the common rail 64, and the common rail 64 stores the pumped fuel at a high fuel pressure. When the direct injection injector 67 is opened, the fuel stored in the common rail 64 is injected from the injection port of the direct injection injector 67. Here, although not shown, the high-pressure fuel pump 63 is a plunger type pump, and is driven by the engine 1 by being connected to a timing belt between a crankshaft and a camshaft, for example. The high-pressure fuel supply system 62 configured to include this engine-driven pump makes it possible to supply fuel with a high fuel pressure of 40 MPa or more to the direct injection injector 67. The pressure of the fuel supplied to the direct injection injector 67 is changed according to the operating state of the engine 1 as will be described later. The high-pressure fuel supply system 62 is not limited to this.

  Similarly, the fuel tank (not shown) and the port injector 68 are connected to each other by a low-pressure fuel supply path. On this low pressure fuel supply path, a low pressure fuel supply system 66 for supplying fuel with a relatively low fuel pressure to the port injector 68 is interposed. Although not shown in detail, the low-pressure fuel supply system 66 includes an electric or engine-driven low-pressure fuel pump and a regulator, and is configured to supply a predetermined pressure of fuel to each port injector 68. . Since the port injector 68 injects fuel into the intake port, the pressure of the fuel supplied by the low pressure fuel supply system 66 is set lower than the pressure of the fuel supplied by the high pressure fuel supply system 62.

  A spark plug 25 that ignites the air-fuel mixture in the combustion chamber 19 is also attached to the cylinder head 12. The spark plug 25 is disposed through the cylinder head 12 so as to extend obliquely downward from the exhaust side of the engine 1. As shown in FIG. 3, the front end of the spark plug 25 is disposed facing the combustion chamber 19 in the vicinity of the front end of the direct injection injector 67 disposed in the center portion of the combustion chamber 19.

  An intake passage 30 is connected to one side of the engine 1 so as to communicate with the intake port 16 of each cylinder 18. On the other hand, an exhaust passage 40 for discharging burned gas (exhaust gas) from the combustion chamber 19 of each cylinder 18 is connected to the other side of the engine 1.

  An air cleaner 31 that filters intake air is disposed at the upstream end of the intake passage 30. A surge tank 33 is disposed near the downstream end of the intake passage 30. The intake passage 30 downstream of the surge tank 33 is an independent passage branched for each cylinder 18, and the downstream end of each independent passage is connected to the intake port 16 of each cylinder 18.

  Between the air cleaner 31 and the surge tank 33 in the intake passage 30, a water-cooled intercooler / warmer 34 that cools or heats the air and a throttle valve 36 that adjusts the amount of intake air to each cylinder 18 are arranged. It is installed. An intercooler bypass passage 35 that bypasses the intercooler / warmer 34 is also connected to the intake passage 30. The intercooler bypass passage 35 is connected to an intercooler for adjusting the flow rate of air passing through the passage 35. A bypass valve 351 is provided. The temperature of fresh air introduced into the cylinder 18 is adjusted by adjusting the ratio between the flow rate of the intercooler bypass passage 35 and the flow rate of the intercooler / warmer 34 through the adjustment of the opening degree of the intercooler bypass valve 351.

  The upstream portion of the exhaust passage 40 is constituted by an exhaust manifold having an independent passage branched for each cylinder 18 and connected to the outer end of the exhaust port 17 and a collecting portion where the independent passages gather. . A direct catalyst 41 and an underfoot catalyst 42 are connected downstream of the exhaust manifold in the exhaust passage 40 as exhaust purification devices for purifying harmful components in the exhaust gas. Each of the direct catalyst 41 and the underfoot catalyst 42 includes a cylindrical case and, for example, a three-way catalyst disposed in a flow path in the case.

  A portion between the surge tank 33 and the throttle valve 36 in the intake passage 30 and a portion upstream of the direct catalyst 41 in the exhaust passage 40 are used for returning a part of the exhaust gas to the intake passage 30. They are connected via a passage 50. The EGR passage 50 includes a main passage 51 in which an EGR cooler 52 for cooling the exhaust gas with engine coolant is disposed, and an EGR cooler bypass passage 53 for bypassing the EGR cooler 52. ing. The main passage 51 is provided with an EGR valve 511 for adjusting the amount of exhaust gas recirculated to the intake passage 30, and the EGR cooler bypass passage 53 has a flow rate of exhaust gas flowing through the EGR cooler bypass passage 53. An EGR cooler bypass valve 531 for adjustment is provided.

  The engine 1 configured as described above is controlled by a powertrain control module (hereinafter referred to as PCM) 10. The PCM 10 includes a microprocessor having a CPU, a memory, a counter timer group, an interface, and a path connecting these units. This PCM 10 constitutes a controller.

As shown in FIGS. 1 and 2, detection signals from various sensors SW <b> 1 to SW <b> 16 are input to the PCM 10. The various sensors include the following sensors. That is, the air flow sensor SW1 that detects the flow rate of fresh air, the intake air temperature sensor SW2 that detects the temperature of fresh air, the downstream side of the intercooler / warmer 34, and the intercooler / warmer 34 downstream of the air cleaner 31. A second intake air temperature sensor SW3 for detecting the temperature of fresh air after passing through the EGR gas temperature sensor SW4, which is disposed in the vicinity of the connection portion of the EGR passage 50 with the intake passage 30 and detects the temperature of the external EGR gas. An intake port temperature sensor SW5 that is attached to the intake port 16 and detects the temperature of the intake air just before flowing into the cylinder 18, and an in-cylinder pressure sensor SW6 that is attached to the cylinder head 12 and detects the pressure in the cylinder 18. The exhaust passage 40 is disposed in the vicinity of the connection portion of the EGR passage 50, and the exhaust temperature and the exhaust respectively. Exhaust temperature sensor SW7 and exhaust pressure sensor SW8 for detecting a force, and is disposed on the upstream side of the direct catalyst 41, the linear O ₂ sensor SW9, direct catalyst 41 and underfoot catalyst 42 for detecting the oxygen concentration in the exhaust gas The lambda O ₂ sensor SW10 that detects the oxygen concentration in the exhaust gas, the water temperature sensor SW11 that detects the temperature of the engine coolant, the crank angle sensor SW12 that detects the rotation angle of the crankshaft 15, and the vehicle An accelerator opening sensor SW13 for detecting an accelerator opening corresponding to an operation amount of an accelerator pedal (not shown), intake side and exhaust side cam angle sensors SW14 and SW15, and a common rail 64 of the high pressure fuel supply system 62 are attached. Further, a fuel pressure sensor SW for detecting the fuel pressure supplied to the direct injection injector 67 16.

  The PCM 10 performs various calculations based on these detection signals to determine the state of the engine 1 and the vehicle, and in response to this, the direct injection injector 67, the port injector 68, the spark plug 25, the VVT 72 on the intake valve side, Control signals are output to actuators of the CVVL 73, the VVL 71 on the exhaust valve side, the high-pressure fuel supply system 62, and various valves (throttle valve 36, intercooler bypass valve 351, EGR valve 511, and EGR cooler bypass valve 531). Thus, the PCM 10 operates the engine 1.

  FIG. 4 shows an example of the operation region of the engine 1. In order to improve fuel efficiency and exhaust emission, the engine 1 performs compression ignition combustion in which combustion is performed by compression self-ignition without performing ignition by the spark plug 25 in a low load region where the engine load is relatively low. Do. However, as the load on the engine 1 increases, in the compression ignition combustion, the combustion becomes too steep and causes problems such as combustion noise. Therefore, in the engine 1, in a high load region where the engine load is relatively high, the compression ignition combustion is stopped, and the engine 1 is switched to the spark ignition combustion using the spark plug 25. Further, the reaction time becomes short as the rotational speed of the engine 1 increases, and compression ignition becomes difficult or compression ignition does not occur. Therefore, in the engine 1, spark ignition combustion is performed in the high speed range even in a relatively low load range. Therefore, the engine 1 is configured to switch between a CI (Compression Ignition) mode in which compression ignition combustion is performed and an SI (Spark Ignition) mode in which spark ignition combustion is performed according to the operating state of the engine 1. The boundary line for switching the combustion mode between the CI mode and the SI mode is set to the lower right in the map of engine speed and engine load shown in FIG. However, the boundary line for mode switching is not limited to the illustrated example.

  As will be described in detail later, in the CI mode, for example, a relatively homogeneous lean mixing is achieved by injecting fuel into the cylinder 18 by the direct injection injector 67 at a relatively early timing, for example, during an intake stroke or a compression stroke. The air-fuel mixture is compressed and self-ignited in the vicinity of the compression top dead center. In contrast, in the SI mode, basically, for example, during the intake stroke or the compression stroke, the direct injection injector 67 injects fuel into the cylinder 18 to form a homogeneous or stratified air-fuel mixture and improve the compression. The mixture is ignited by performing ignition in the vicinity of the dead center. In the SI mode, the engine 1 is also operated at the theoretical air fuel ratio (λ = 1). This makes it possible to use a three-way catalyst, which is advantageous for improving the emission performance.

  The geometric compression ratio of the engine 1 is set to 14 or more (for example, 18) as described above. Since the high compression ratio increases the compression end temperature and the compression end pressure, the CI mode is advantageous in stabilizing the compression ignition combustion. On the other hand, since the high compression ratio engine 1 is switched to the SI mode in the high load range, the higher the engine load, particularly in the low speed range, the more likely abnormal combustion such as pre-ignition and knocking is likely to occur. There is an inconvenience (see the white arrow in FIG. 4).

  Therefore, in this engine 1, when the engine operating state is in a high load region in the low speed region, abnormal combustion is avoided by executing SI combustion in which the fuel injection form is greatly different from the conventional one. . Specifically, this fuel injection mode is a period that is significantly retarded from the latter half of the compression stroke to the early stage of the expansion stroke with a fuel pressure that is significantly higher than in the past (hereinafter, this period is referred to as the retard period). The fuel is injected into the cylinder 18 by the direct injection injector 67. This characteristic fuel injection mode is hereinafter referred to as “high pressure retarded injection”.

  FIG. 5 shows the heat generation rate (upper figure) and the progress of the unburned mixture reaction in the SI combustion (solid line) by the high-pressure retarded injection described above and the conventional SI combustion (broken line) in which fuel injection is performed during the intake stroke. It is a figure which compares the difference in a degree (lower figure). The horizontal axis in FIG. 5 is the crank angle. As a premise for this comparison, the operating state of the engine 1 is a high load region in the low speed region, and the amount of fuel to be injected is the same in the case of SI combustion by high pressure retarded injection and conventional SI combustion.

  First, in the conventional SI combustion, a predetermined amount of fuel is injected into the cylinder 18 during the intake stroke (broken line in the upper diagram). In the cylinder 18, a relatively homogeneous air-fuel mixture is formed after the fuel injection until the piston 14 reaches the compression top dead center. In this example, ignition is executed at a predetermined timing indicated by a white circle after the compression top dead center, thereby starting combustion. After the start of combustion, as shown by the broken line in the upper diagram of FIG. Here, the period from the start of fuel injection to the end of combustion corresponds to the reaction possible time of the unburned mixture (hereinafter sometimes simply referred to as the reaction possible time). During this time, the reaction of the unburned mixture gradually proceeds. The dotted line in the figure shows the ignition threshold, which is the degree of reactivity with which the unburned mixture reaches ignition. Conventional SI combustion has a very long reaction time, during which the unburned mixture reacts. Since it continues to advance, the reactivity of the unburned mixture before and after ignition exceeds the ignition threshold, causing abnormal combustion such as premature ignition or knocking.

  On the other hand, the high pressure retarded injection aims to shorten the reaction possible time, thereby avoiding abnormal combustion. That is, as shown in FIG. 5, the reaction possible time is the period during which the direct injection injector 67 injects fuel ((1) injection period) and after the end of the injection, the combustible air-fuel mixture is around the spark plug 25. The sum of the period until formation ((2) mixture formation period) and the period until combustion ended by ignition ((3) combustion period), that is, (1) + (2) + (3). The high-pressure retarded injection shortens the injection period, the mixture formation period, and the combustion period, thereby shortening the reaction time. This will be described in order.

  First, the high fuel pressure relatively increases the amount of fuel injected from the direct injection injector 67 per unit time. For this reason, as shown in (1) in the middle of FIG. 6, when the fuel injection amount is constant, the relationship between the fuel pressure and the fuel injection period is generally longer as the fuel pressure is lower. Thus, the higher the fuel pressure, the shorter the injection period. Therefore, the high pressure retarded injection in which the fuel pressure is set to be significantly higher than the conventional one shortens the injection period.

  Further, the high fuel pressure is advantageous for atomization of the fuel spray injected into the cylinder 18 and makes the flight distance of the fuel spray longer. For this reason, as shown in FIG. 6A in the lower part of FIG. 6, the relationship between the fuel pressure and the fuel evaporation time is approximately, the lower the fuel pressure, the longer the fuel evaporation time, and the higher the fuel pressure, the fuel evaporation time. Becomes shorter. In addition, as shown in the lower part of FIG. 6 by (B), the time until the fuel spray reaches the fuel pressure and the spark plug 25 is generally longer. The lower the fuel pressure, the longer the time to reach, The higher the fuel pressure, the shorter the time to reach. The time until the fuel spray reaches around the spark plug 25 can be calculated from the spray flight distance from the tip of the direct injection injector 67 to the spark plug 25 and the fuel injection speed proportional to the fuel pressure. . The mixture formation period is a time ((A) + (B)) obtained by adding the fuel evaporation time and the fuel spray arrival time around the spark plug 25. As shown in the figure, the higher the fuel pressure, the shorter the mixture formation period. Therefore, the high pressure retarded injection in which the fuel pressure is set to be significantly higher than the conventional case shortens the mixture formation period as a result of the fuel evaporation time and the fuel spray arrival time around the spark plug 25 being reduced. On the other hand, as shown by white circles in the figure, the conventional intake stroke injection at a low fuel pressure significantly increases the mixture formation period. As described above, the combination of the multi-hole injector 67 and the cavity 141 shortens the time until the fuel spray reaches around the spark plug 25 after the fuel injection, resulting in the mixture formation period. Effective for shortening.

  Thus, shortening the injection period and the mixture formation period makes it possible to set the fuel injection timing, more precisely, the injection start timing to a relatively late timing. Therefore, in the high pressure retarded injection, as shown in the upper diagram of FIG. 5, the fuel is injected within the retard period from the latter stage of the compression stroke to the early stage of the expansion stroke. As the fuel is injected into the cylinder 18 at a high fuel pressure, the turbulence in the cylinder becomes stronger and the turbulence energy in the cylinder 18 increases. This high turbulence energy is a timing at which the fuel injection timing is relatively late. Therefore, it is advantageous for shortening the combustion period.

  That is, when the fuel injection is performed within the retard period as shown in (D) in the lower part of FIG. 6, the relationship between the fuel pressure and the turbulent energy within the combustion period is generally lower as the fuel pressure is lower. The turbulent energy increases as the turbulent energy decreases and the fuel pressure increases. In addition, the line shown with a broken line in the same figure is an example at the time of performing fuel injection during an intake stroke. Even if fuel is injected into the cylinder 18 at a high fuel pressure, if the injection timing is in the intake stroke, the time until the ignition timing is long, or the inside of the cylinder 18 is compressed in the compression stroke after the intake stroke. As a result, the disturbance in the cylinder 18 is attenuated. As a result, when fuel is injected during the intake stroke, the turbulent energy during the combustion period becomes relatively low regardless of the fuel pressure level.

  As shown in FIG. 6 (C) in the lower part of FIG. 6, the relationship between the turbulent energy and the combustion period in the combustion period is generally such that the lower the turbulent energy, the longer the combustion period, and the higher the turbulent energy, the more combustion occurs. The period is shortened. Accordingly, from (C) and (D) of FIG. 6, the relationship between the fuel pressure and the combustion period is as shown in (3) in the middle of FIG. The higher the pressure, the shorter the combustion period. That is, the high pressure retarded injection shortens the combustion period. On the other hand, as indicated by white circles in the figure, the conventional intake stroke injection at a low fuel pressure has a long combustion period. The multi-injector type injector 67 is advantageous for improving the turbulent energy in the cylinder 18 and is effective for shortening the combustion period. In addition, the combination of the multi-injector type injector 67 and the cavity 141 provides fuel. Putting the spray in the cavity 141 is also effective for shortening the combustion period.

  From the relationship between the fuel pressure and the combustion period shown in FIG. 6 (3), in other words, from the curved shape, the combustion period is effectively shortened by setting the fuel pressure to 40 MPa or more, for example. Is possible. Moreover, the fuel pressure of 40 MPa or more can effectively shorten both the injection period and the mixture formation period. The fuel pressure is preferably set as appropriate according to the properties of the fuel used, which contains at least gasoline. The upper limit may be 120 MPa as an example.

  As described above, the high pressure retarded injection shortens the injection period, the mixture formation period, and the combustion period, and as a result, as shown in FIG. 5, the fuel injection start timing SOI to the combustion end timing θend are not changed. It is possible to significantly shorten the reaction time of the fuel mixture as compared with the case of fuel injection during the conventional intake stroke. As a result of shortening this reaction possible time, as shown in the upper diagram of FIG. 6, in the conventional intake stroke injection at a low fuel pressure, as shown by a white circle, the reaction progress of the unburned mixture at the end of combustion is shown. However, when the ignition threshold is exceeded and abnormal combustion occurs, the high-pressure retarded injection suppresses the progress of the reaction of the unburned mixture at the end of combustion, as shown by the black circle, to prevent abnormal combustion. It can be avoided. It should be noted that the ignition timing is set to the same timing in the white circle and the black circle in the upper diagram of FIG.

  High pressure retarded injection avoids abnormal combustion by devising the form of fuel injection into the cylinder 18. Unlike this, it is conventionally known that the ignition timing is retarded for the purpose of avoiding abnormal combustion. The retarding of the ignition timing suppresses the progress of the reaction by suppressing the increase in the temperature and pressure of the unburned mixture. FIG. 7 shows the relationship between the ignition timing and the reaction progress of the unburned mixture at the end of combustion. The broken line in the figure is the case of SI combustion in which conventional intake stroke injection is performed, and the solid line is the case of SI combustion in which high pressure retarded injection is performed. As described above, since the retarding of the ignition timing suppresses the progress of the reaction of the unburned mixture, the solid line and the broken line are lowered to the right. In addition, as described above, the high-pressure retarded injection suppresses the progress of the reaction of the unburned mixture by the fuel injection. Therefore, when compared at the same ignition timing, the conventional SI combustion that performs the intake stroke injection is the high-pressure retarded injection. The reaction of the unburned mixture proceeds more than the SI combustion that performs injection. That is, the broken line is positioned above the solid line. For this reason, when performing the conventional intake stroke injection (white circle), if the ignition timing is not retarded as compared with the case of performing high pressure retarded injection (black circle), the reaction progress of the unburned mixture will reach the ignition threshold. It will be over. In other words, when performing high pressure retarded injection, it is possible to advance the ignition timing more than when performing conventional intake stroke injection.

  FIG. 8 is a diagram showing the relationship between ignition timing, thermal efficiency, and torque. The ignition timing at which the thermal efficiency and torque are maximized is near the compression top dead center, and the thermal efficiency and torque decrease as the ignition timing is retarded. As described above, when performing the intake stroke injection, the ignition timing must be delayed as shown by a white circle, whereas when performing high pressure retarded injection, the ignition timing is advanced as shown by the black circle. Since it is possible to approach the compression top dead center, the thermal efficiency and torque are improved. That is, the high pressure retarded injection not only avoids abnormal combustion, but also makes it possible to advance the ignition timing by the amount that can be avoided, which is advantageous in improving fuel consumption.

  Here, the characteristics of SI combustion by high-pressure retarded injection will be briefly described with reference to FIG. FIG. 9A is a diagram showing a change in heat generation rate (dQ / dθ) with respect to the crank angle, and FIG. 9B is a diagram showing a change in in-cylinder pressure increase rate (dP / dθ) with respect to the crank angle. The solid line in the figure indicates the case where SI combustion is performed by high pressure retarded injection, and the broken line in the figure indicates the case where compression ignition combustion (CI combustion) is performed. The operating state of the engine 1 is in a high load range in the low speed range. First, in CI combustion, as shown in FIG. 5A, the combustion becomes steep and the combustion period becomes extremely short. In addition, as shown in FIG. 5B, the peak of the in-cylinder pressure becomes too high and exceeds the allowable value, causing a problem of combustion noise. That is, this indicates that CI combustion cannot be performed when the operating state of the engine 1 is in a high load range in the low speed range.

  On the other hand, the SI combustion by the high pressure retarded injection ensures a large heat generation rate and an appropriate combustion period to obtain a sufficient torque as shown in FIG. As shown in FIG. 3, the peak of the in-cylinder pressure becomes lower than the allowable value, and the generation of combustion noise can be avoided. That is, when the operating state of the engine 1 is in a high load range in the low speed range, SI combustion by high pressure retarded injection is extremely effective.

  Next, with reference to FIG. 10, an operation state of the intake valve 21 and the exhaust valve 22 corresponding to the operation state of the engine 1 and a control example of the fuel injection timing and the ignition timing will be described. Here, each of (a), (b), (c), and (d) in FIG. 10 basically has the operating state of the engine 1 in the low speed range, and (a) <(b) <(c) <( The engine load increases in the order of d). (A) (b) is a low load region corresponding to the CI mode, and (c) is a high load region corresponding to the SI mode. (D) is a fully open load region corresponding to the SI mode. Note that (d) also corresponds to the case where the operating range of the engine 1 is in the medium speed range within the high load range.

  First, FIG. 10A shows a case where the operating state of the engine 1 is in a low load region within a low speed region. Since this operation region is in the CI mode, the exhaust valve 22 is opened twice during the intake stroke under the control of the VVL 71 (see the solid line Ex2 in the figure. The solid line is the exhaust valve 22). ), And the broken line indicates the lift curve of the intake valve 21), whereby the internal EGR gas is introduced into the cylinder 18. The introduction of internal EGR gas increases the compression end temperature and stabilizes compression ignition combustion. The fuel injection timing is set during the intake stroke, and the direct injection injector 67 injects fuel into the cylinder 18 to form a homogeneous lean air-fuel mixture in the cylinder 18. The fuel injection amount is set according to the load of the engine 1.

  FIG. 10B also shows a case where the operating state of the engine 1 is in a low load region within a low speed region. However, the engine load in FIG. 10B is higher than that in FIG. Since this operating region is also in the CI mode, the exhaust is opened twice by the control of the VVL 71 and the internal EGR gas is introduced into the cylinder 18 as described above. However, since the temperature in the cylinder 18 naturally increases as the engine load increases, the internal EGR amount is reduced from the viewpoint of avoiding premature ignition. As illustrated in FIG. 10, the internal EGR amount may be adjusted by adjusting the lift amount of the intake valve 21 under the control of the CVVL 73. Although not shown in FIG. 10, the internal EGR amount may be adjusted by adjusting the opening of the throttle valve 36. The fuel injection timing is set to an appropriate timing during the intake stroke or the compression stroke. At this timing, the direct injection injector 67 injects fuel into the cylinder 18 to form a homogeneous or stratified lean mixture in the cylinder 18. The point that the fuel injection amount is set according to the load of the engine 1 is the same as in FIG.

  10A and 10B show an example in which the valve opening period of the exhaust valve 22 during the intake stroke is set in the first half of the intake stroke. The valve opening period of the exhaust valve 22 may be set in the latter half of the intake stroke. When the valve opening period is set to the first half of the intake stroke, the exhaust valve 22 may be left open from the exhaust stroke sandwiching the exhaust top dead center to the first half of the intake stroke.

  FIG. 10C shows a case where the operating state of the engine 1 is in a high load region within a low speed region. This operation region is the SI mode, and in this operation region, the opening of the exhaust valve 22 is stopped twice. In the SI mode, the filling amount is adjusted so that the air-fuel ratio λ = 1. The adjustment of the filling amount may be performed by slow closing of the intake valve 21 in which the throttle valve 36 is fully opened while the closing timing of the intake valve 21 is set after the intake bottom dead center by the control of the VVT 72 and the CVVL 73. Good. This is advantageous for reducing pump loss. The adjustment of the filling amount may also be performed by adjusting the opening amount of the EGR valve 511 and adjusting the amount of fresh air introduced into the cylinder 18 and the amount of external EGR gas while the throttle valve 36 is fully opened. . This is effective for reducing pumping loss and cooling loss. In addition, introduction of external EGR gas contributes to avoiding abnormal combustion and also has an advantage of suppressing generation of Raw NOx. Further, as the adjustment of the filling amount, the slow closing control of the intake valve 21 and the control of the external EGR may be combined. In particular, on the low load side in the high load region, in order to prevent the EGR rate from being too high, the charging amount is adjusted by the slow closing control of the intake valve 21 while introducing the external EGR into the cylinder 18. Also good.

  The form of fuel injection is the high-pressure retarded injection described above. Therefore, the direct injection injector 67 directly injects the fuel into the cylinder 18 with a high fuel pressure during the retard period from the latter half of the compression stroke to the early stage of the expansion stroke. The high-pressure retarded injection may be configured by one injection (that is, batch injection), but as shown in FIG. 10C, two injections of the first injection and the subsequent second injection are performed. It may be configured to perform within the retard period (that is, divided injection). Since the first injection can ensure a relatively long air-fuel mixture formation period, it is advantageous for fuel vaporization atomization. Since a sufficient mixture formation period is secured by the first injection, the injection timing of the second injection can be set to a more retarded timing. This is advantageous for improving the turbulent energy in the cylinder and for shortening the combustion period. When performing split injection, it is preferable to set the fuel injection amount of the second injection to be larger than the fuel injection amount of the first injection. By doing so, the turbulent energy in the cylinder 18 is sufficiently increased, which is advantageous for shortening the combustion period and for avoiding abnormal combustion. Such split injection is performed only on the relatively high load side where the fuel injection amount increases in the high load region, and batch injection is performed on the low load side in the high load region where the fuel injection amount is relatively small. It may be. Further, the number of divisions is not limited to two, and may be set to three or more.

  Thus, in the SI mode, ignition by the spark plug 25 is executed in the vicinity of the compression top dead center after the end of fuel injection.

  FIG. 10 (d) shows a case where the operating state of the engine 1 is in the fully open load region in the low speed region. This operation region is the SI mode as in FIG. 10C, and the exhaust valve 22 is not opened twice. Further, since it is in the fully open load region, the external EGR is also stopped by closing the EGR valve 511.

  The form of fuel injection is basically high-pressure retarded injection, and is constituted by two injections into the cylinder 18 during the retard period of the first injection and the second injection, as shown in the figure. . The high pressure retarded injection may be batch injection. In this fully open load region, injection during the intake stroke may be added for the purpose of improving the intake charge efficiency. In this intake stroke injection, the intake air charging efficiency is improved by the intake air cooling effect accompanying the fuel injection, which is advantageous in improving the torque. Accordingly, when the operating state of the engine 1 is in the fully open load region in the low speed region, three fuel injections of the intake stroke injection and the first and second injections are executed, or the intake stroke injection and the batch Two fuel injections with injection are performed.

  Here, as described above, the high pressure retarded injection in which the fuel is directly injected into the cylinder 18 by the direct injection injector 67 has a very high fuel pressure. Therefore, if fuel is directly injected into the cylinder 18 during the intake stroke with such a high fuel pressure, a large amount of fuel may adhere to the wall surface in the cylinder 18 and cause problems such as oil dilution. There is. Therefore, in this intake stroke injection, fuel is injected into the intake port 16 not through the direct injection injector 67 but through the port injector 68 that injects fuel with a relatively low fuel pressure. By doing so, the above-mentioned problems such as oil dilution can be avoided.

  Further, as described above, FIG. 10D corresponds to a case where the operation region of the engine 1 is in the medium speed region within the high load region. When the operating state of the engine 1 is in the middle speed range, the flow in the cylinder 18 is stronger than in the low speed range, and the actual time for changes in the crank angle is shortened, which is advantageous in avoiding abnormal combustion. Therefore, abnormal combustion can be avoided even if the injection amount of the high-pressure retarded injection performed within the retard period from the late stage of the compression stroke to the early stage of the expansion stroke is reduced. Therefore, when the operating state of the engine 1 is in the medium speed range in the high load range, the fuel injection amount of the high pressure retarded injection is reduced, and the amount is allocated to the intake stroke injection that is injected during the intake stroke. As a result, as described above, the intake charge efficiency is improved, and as a result, torque is improved. Therefore, when the operating state of the engine 1 is in the medium speed range in the high load range, both avoiding abnormal combustion and improving the torque are compatible. It should be noted that when the operating state of the engine 1 is in a high load range in the low speed range (specifically, a fully open load range) and in a medium speed range in the high load range, in other words, in the high load range. When the low speed region and the medium speed region are compared, the fuel injection amount of the intake stroke injection may be increased when in the medium speed region than when in the low speed region.

  11 to 14 show the parameters of the engine 1 with respect to the load fluctuation in the low speed range, that is, (b) the opening degree of the throttle valve 36, (c) the opening degree of the EGR valve 511, and (d) the exhaust valve 22. The control examples of the opening timing of the opening degree, (e) the opening timing of the intake valve 21, (f) the closing timing of the intake valve 21, and (g) the lift amount of the intake valve are shown.

  FIG. 11A shows the state in the cylinder 18. This figure shows the configuration of the air-fuel mixture in the cylinder with the horizontal axis representing torque (in other words, engine load) and the vertical axis representing the air-fuel mixture charge amount in the cylinder. As described above, the area on the left side of the figure where the load is relatively low is the CI mode, and the area on the right side of the figure where the load is higher than the predetermined load is the SI mode. The fuel amount (total fuel amount) is increased as the load increases regardless of the CI mode and the SI mode. A fresh air amount for setting the stoichiometric air-fuel ratio (λ = 1) is set with respect to this fuel amount, and this new air amount increases with an increase in the fuel amount as the load increases. It will be.

  In the CI mode, as described above, since the internal EGR gas is introduced into the cylinder 18, the remaining amount of the filling amount is constituted by the internal EGR gas and excess fresh air. Therefore, in the CI mode, a lean mixture is obtained.

  On the other hand, in the SI mode, the engine 1 is operated so that λ = 1, and the introduction of the internal EGR gas is stopped. As an example of control, in FIG. 11, the filling amount into the cylinder 18 in the SI mode is reduced. In FIG. 11, in particular, in the SI mode, by adjusting the valve closing timing of the intake valve 21. , Control the filling amount.

  As shown in FIG. 11B, the throttle valve 36 is set to be fully opened regardless of the load level of the engine 1 so that the state in the cylinder 18 is as shown in FIG. The EGR valve 511 is set to fully closed regardless of the load of the engine 1 as shown in FIG. This control is advantageous in reducing pump loss.

  FIG. 11D shows the valve closing timing when the exhaust valve 22 is opened twice. In the CI mode, as described above, the valve closing timing is set to a predetermined timing between the exhaust top dead center and the intake bottom dead center in order to introduce the internal EGR gas into the cylinder 18. On the other hand, in the SI mode, the valve closing timing is set to the exhaust top dead center. That is, in the SI mode, the control of the internal EGR is stopped as a result of stopping the exhaust opening twice.

  In this way, with respect to the double opening of the exhaust valve 22 set at a predetermined valve closing timing in the CI mode, as shown in FIG. The higher the value is, the closer it is to the exhaust top dead center. Therefore, the internal EGR gas introduced into the cylinder 18 increases as the load on the engine 1 decreases, whereas the internal EGR gas introduced into the cylinder 18 decreases as the load on the engine 1 increases. . As the load on the engine 1 is lower, the compression end temperature in the cylinder 18 is increased by a large amount of internal EGR gas, which is advantageous in realizing stable compression ignition combustion. On the other hand, the higher the load of the engine 1, the lower the compression end temperature in the cylinder 18 by suppressing the internal EGR gas, which is advantageous in suppressing premature ignition. In the SI mode, the opening timing of the intake valve 21 is advanced further than the exhaust top dead center, and the advance amount increases as the engine load increases.

  On the other hand, as shown in FIG. 11 (f), the closing timing of the intake valve 21 is made constant at the intake bottom dead center in the CI mode. On the other hand, in the SI mode, the closing timing of the intake valve 21 is retarded from the intake bottom dead center. The retard amount increases gradually when the engine load increases so that the retard amount increases when the load of the engine 1 is relatively low and decreases when the load of the engine 1 is relatively high. Is set to be Thus, in the SI mode, the filling amount is reduced by the slow closing control of the intake valve 21. Instead of the slow closing control of the intake valve 21, the filling amount may be reduced by adjusting the opening degree of the throttle valve 36 in the SI mode.

  Further, as shown in FIG. 11 (g), the lift amount of the intake valve 21 gradually increases from the minimum lift amount in the CI mode as the engine load increases, whereas in the SI mode, the engine load increases. Regardless of the height, the maximum lift amount is set constant.

  FIG. 12 shows a control example different from FIG. 11, and FIGS. 11 and 12 show the opening degree (b) of the throttle valve 36, the opening timing (e) of the intake valve 21 in the CI mode, Control of the lift amount (g) of the intake valve 21 is different from each other. That is, in the control shown in FIG. 12, first, as shown in FIG. 12B, the throttle valve 36 is throttled in the CI mode, and the opening of the throttle valve 36 is small on the low load side in the CI mode. Control is performed so that the engine load gradually increases as the engine load increases so as to increase on the high load side. On the other hand, in the SI mode, the throttle valve 36 is fully opened.

  Further, as shown in FIG. 12 (e), in the CI mode, the opening timing of the intake valve 21 is made constant at the exhaust top dead center regardless of the engine load level, and in FIG. 12 (g). As shown, in the CI mode, the lift amount of the intake valve 21 is made constant at a predetermined lift amount regardless of the engine load level. By the combination of the control of the throttle valve 36 and the control of the intake valve 21, the internal EGR gas amount introduced into the cylinder 18 is adjusted according to the opening degree of the throttle valve 36 in the CI mode. become. Therefore, as apparent from a comparison between FIG. 11A and FIG. 12A, in the control example shown in FIG. 12, the configuration of the air-fuel mixture filled in the cylinder 18 is the same as the control shown in FIG. Same as example.

  FIG. 13 shows a control example different from FIG. 11 (and FIG. 12). FIGS. 11 and 13 show the opening degree (c) of the EGR valve 511 and the opening of the intake valve 21 in the SI mode. Control of timing (e) and closing timing (f) of the intake valve 21 are different from each other. That is, in FIG. 11 (and FIG. 12), as shown in FIG. 11A, the filling amount into the cylinder 18 is reduced by adjusting the valve closing timing of the intake valve 21 in the SI mode. On the other hand, in the control example of FIG. 13, as shown in FIG. 13A, external EGR gas is introduced into the cylinder 18 in the SI mode.

  First, as shown in FIG. 13C, the EGR valve 511 remains closed in the CI mode, whereas it is opened in the SI mode. The opening degree of the EGR valve 511 is gradually decreased as the engine load increases so that the lower the load and the higher the load in the SI mode. More precisely, it is fully open at the switching between the CI mode and the SI mode, and is fully closed at the fully open load. Therefore, in this control example, even in the SI mode, the external EGR gas is not introduced into the cylinder 18 at the fully open load.

  Further, as shown in FIG. 13 (e), the opening timing of the intake valve 21 is made constant at the exhaust top dead center in the SI mode, and the intake valve 21 is closed as shown in FIG. 13 (f). The timing is constant at the intake bottom dead center in the SI mode. Accordingly, in the SI mode, the throttle valve 36 is made constant when fully open (FIG. 13B), the opening timing and closing timing of the intake valve 21 are made constant, and the lift amount is made constant at the maximum. (FIG. 13 (g)). From this, the ratio between the amount of fresh air introduced into the cylinder 18 and the amount of external EGR gas is adjusted by adjusting the opening of the EGR valve 511. Such control is advantageous in reducing pump loss. In addition, in the SI mode, introducing the external EGR gas into the cylinder 18 is advantageous in reducing cooling loss, avoiding abnormal combustion, and suppressing Raw NOx.

  The control example shown in FIG. 14 is an example in which the control example shown in FIG. 12 is adopted for the CI mode, and the control example shown in FIG.

  Next, FIG. 15 shows each control parameter of the engine 1 with respect to the load fluctuation in the low speed range, that is, (b) G / F, (c) injection timing, (d) fuel pressure, (e) fuel injection pulse width ( That is, the injection period) and (f) changes in the ignition timing are shown.

  First, in the case of following the control example of FIG. 11 or FIG. 12, the state of the air-fuel mixture in the cylinder is configured as shown in FIG. For this reason, as shown in FIG. 15B, G / F gradually approaches the stoichiometric air-fuel ratio from lean as the fuel amount increases in the CI mode. On the other hand, in the SI mode, as described above, since the filling amount is decreased, G / F is constant at the theoretical air-fuel ratio (G / F = 14.7).

  As shown in FIG. 15C, the fuel injection timing is set, for example, during the intake stroke between the exhaust top dead center and the intake bottom dead center in the CI mode. The fuel injection timing may be changed according to the load of the engine 1. On the other hand, in the SI mode, the fuel injection timing is set to a retard period from the latter half of the compression stroke to the early stage of the expansion stroke. That is, high pressure retarded injection. In the SI mode, the injection timing is gradually changed to the retard side as the engine load increases. This is because the pressure and temperature in the cylinder 18 increases as the engine load increases, and abnormal combustion is likely to occur. Therefore, in order to avoid this efficiently, the injection timing is set to the retard side. This is necessary. Here, the solid line in FIG. 15C shows an example of the fuel injection timing in the case of batch injection in which the high pressure retarded injection is performed by one fuel injection. On the other hand, the alternate long and short dash line in FIG. 15C indicates the fuel injection of each of the first injection and the second injection when the high-pressure retarded injection is divided into two fuel injections of the first injection and the second injection. An example of timing is shown. According to this, since the second injection in the divided injection is executed on the retard side as compared with the case of performing the batch injection, it is advantageous for avoiding abnormal combustion. This is because, as described above, the first injection is executed relatively early to ensure the vaporization time of the fuel, and the fuel injection amount of the second injection is relatively reduced. This is due to the shortening of the conversion time.

  Further, as indicated by a dotted line in FIG. 15 (c), since the overall fuel injection amount increases in the fully open load region, the intake stroke injection is performed for the purpose of improving the intake charging efficiency by increasing the fuel injection amount. You may make it perform.

  FIG. 15D shows a change in the fuel pressure supplied to the direct injection injector 67. In the CI mode, the fuel pressure is set constant at the minimum fuel pressure. In contrast, in the SI mode, the fuel pressure is set to be higher than the minimum fuel pressure, and the fuel pressure is set to increase as the engine load increases. This is because abnormal combustion tends to occur as the engine load increases, so that further shortening of the injection period and further retarding of the injection timing are required.

  FIG. 15 (e) shows the change in the injection pulse width (injector opening period) corresponding to the injection period in the case of batch injection. In the CI mode, the pulse width also increases with the fuel injection amount. Similarly, in the SI mode, the pulse width increases as the fuel injection amount increases. However, as shown in FIG. 4D, the fuel pressure in the SI mode is set to be significantly higher than that in the CI mode, so that the fuel injection amount in the SI mode is larger than the fuel injection amount in the CI mode. Nevertheless, the pulse width is set shorter than the pulse width in the CI mode. This shortens the unburned mixture reaction possible time and is advantageous for avoiding abnormal combustion.

  FIG. 15 (f) shows the change in the ignition timing. In the SI mode, as the fuel injection timing is retarded as the engine load increases, the ignition timing also retards as the engine load increases. Is done. This is advantageous for avoiding abnormal combustion. In the CI mode, although ignition is basically not performed, for the purpose of avoiding smoldering of the spark plug 25, for example, ignition may be performed in the vicinity of the exhaust top dead center as shown by a one-dot chain line in FIG. .

  FIG. 16 is different from FIG. 15 in that (b) G / F, (c) injection timing, (d) fuel pressure, (e) fuel injection pulse width, and FIG. , (F) shows a change in ignition timing. In comparison between FIG. 15 and FIG. 16, (c), (e), and (f) are the same as each other.

  The state of the air-fuel mixture in the cylinder 18 has a configuration as shown in FIG. 16A, and since the external EGR gas is introduced into the cylinder 18 in the SI mode, the G / F is as shown in FIG. As shown in FIG. 5, the engine mode does not become discontinuous from the CI mode to the SI mode, and gradually decreases as the engine load increases. As described above, since the external EGR gas is introduced in the SI mode, particularly when the operating state of the engine 1 is in the middle load range, there is a possibility that the combustion becomes slow and the combustion period becomes long. Therefore, for the purpose of shortening the combustion period, in this control example, as shown in FIG. 16D, the fuel pressure is set higher than that in FIG. 15D (see the alternate long and short dash line in FIG. 16). Is done.

  11 to 16 described above show the mode change in the steady state, while FIG. 17 switches the mode between the CI mode and the SI mode as the load of the engine 1 changes. It shows a transitional change. That is, FIG. 17 similar to (a) of FIG. 11 and the like shows an example of a transient change in the configuration of the air-fuel mixture in the cylinder in the CI mode and the SI mode, and the left side of FIG. Is the same as (a) of FIG. 11 and the like. However, the horizontal axis of FIG. 17 also shows the passage of time, and when the load of the engine 1 gradually increases (at a constant rate in the illustrated example) and the operation mode shifts from the CI mode to the SI mode, Time elapses from the left to the right in FIG. On the other hand, when the load of the engine 1 gradually decreases (at a constant rate in the illustrated example) and the operation mode shifts from the SI mode to the CI mode, the time elapses from the right to the left in FIG. Thus, a switching mode is separately provided between the CI mode and the SI mode in a transient state when switching modes. This switching mode is a mode for smoothly switching between the CI mode and the SI mode, and performs spark ignition combustion at an air-fuel ratio λ = 1.

  First, the switching mode will be described by taking as an example a case where the load of the engine 1 gradually increases and the operation mode shifts from the CI mode to the SI mode. As described above, in the CI mode, for example, as shown in FIG. 11B, the throttle valve 36 is set to be fully open, and the exhaust valve 22 is opened twice through the control of the VVL 71. The closing timing of the intake valve 21 is made constant at the intake bottom dead center. As a result, in addition to the fuel and the fresh air that satisfy λ = 1, the internal EGR gas and the surplus fresh air are introduced into the cylinder 18, so that a lean air-fuel mixture is formed.

  Then, when the load of the engine 1 gradually increases and the timing for shifting from the CI mode to the switching mode is reached, the exhaust valve 22 is opened twice through the control of the VVL 71, and the introduction of the internal EGR gas is stopped. The On the other hand, since the opening degree of the throttle valve 36 and the closing timing of the intake valve 21 through the control of the VVT 72 do not change abruptly, a large amount is used instead of the internal EGR gas that has been introduced into the cylinder 18 until then. Will be introduced into the cylinder 18. In this state, as shown by a one-dot chain line in FIG. 17, when the fuel injection amount corresponding to the load of the engine 1 is set, the air-fuel ratio in the cylinder 18 becomes lean. In order to satisfy the fuel ratio λ = 1, as shown by a solid line in FIG. 17, an amount of fuel commensurate with a large amount of fresh air introduced into the cylinder 18 is injected. This is because the control related to the filling amount of the engine 1 is mechanical control and low responsiveness, whereas the change of the fuel injection amount is electrical control and high responsiveness. In other words, the stoichiometric air-fuel ratio λ = 1 can be satisfied by increasing control of the fuel injection amount with high responsiveness.

  As a result of the increase in the fuel injection amount, a sudden increase in torque is caused in the switching mode. Therefore, in this switching mode, as will be described in detail later, with a relatively high fuel pressure as in the SI mode described above, and by further retarding the fuel injection timing and ignition timing from the SI mode, Suppresses the generated torque and avoids torque shock. The combustion in this switching mode is spark ignition combustion of high pressure retarded injection with a relatively high fuel pressure, so that the combustion can be stabilized even if the ignition timing is greatly retarded.

  In the switching mode, in addition to the retard control of the fuel injection timing and the ignition timing, control for gradually reducing the filling amount with the passage of time is performed. For this purpose, the control example shown here uses the slow closing control of the intake valve 21 that sets the closing timing of the intake valve 21 after the intake bottom dead center through the control of the VVT 72. Specifically, in the switching mode, the valve closing timing of the intake valve 21 is gradually changed to the retard side. As a result, the filling amount is gradually reduced as shown in FIG. Details of the control in this switching mode will be described later. Here, in order to adjust the filling amount in the switching mode, throttle control of the throttle valve 36 may be used instead of the slow closing control of the intake valve 21. However, since the delayed closing control of the intake valve 21 is relatively higher in control response, it is preferable to employ the delayed closing control of the intake valve 21. The slow closing control of the intake valve 21 and the throttle control of the throttle valve 36 may be used in combination. In particular, as described with reference to FIGS. 13 and 14 and the like, if the external EGR gas is introduced into the cylinder 18 in the SI mode, the external EGR gas is used in the switching mode in combination with the reduction in the filling amount described above. The EGR valve 511 may be opened in order to start the introduction. Thus, starting the introduction of the external EGR gas into the cylinder 18 reduces the amount of decrease in the filling amount due to the above-described slow closing control of the intake valve 21 by the amount of the external EGR gas introduced into the cylinder 18. Therefore, it is possible to end the switching mode at an early stage and promptly shift to the SI mode.

  As the filling amount (in other words, fresh air) decreases, the fuel amount corresponding to λ = 1 also decreases. Therefore, as shown in FIG. 17, in the switching mode, the fuel injection amount is gradually decreased with time.

  If the increased fuel injection amount is equivalent to the load on the engine 1, the switching mode is terminated and the normal SI mode is shifted to. As will be described later, the switching mode may be continued for a predetermined time by performing control based on time.

  Thus, in the switching mode related to the transition from the CI mode to the SI mode, the response of the slow closing control of the intake valve 21 by the control of the VVT 72 or the throttle control of the throttle valve 36 is delayed with respect to the suspension of the internal EGR control. In this mode, a large amount of fresh air is introduced into the cylinder 18 to solve the problem. In other words, the fuel injection amount is temporarily increased, and the sudden increase in torque is suppressed by delaying the fuel injection timing and the ignition timing, and the filling amount is gradually decreased to smoothly shift to the SI mode. Transition.

  On the other hand, when the load of the engine 1 gradually decreases and the operation mode shifts from the SI mode to the CI mode, as described above, in the SI mode, as shown in FIG. In order to reduce the filling amount as the load 1 decreases, the closing timing of the intake valve 21 is gradually retarded, while in the CI mode, the closing timing of the intake valve 21 is set to the intake bottom dead center. Therefore, the valve closing timing is greatly changed discontinuously. Therefore, the control responsiveness of the VVT 72 that greatly changes the closing timing of the intake valve 21 with respect to the mode change is delayed. Therefore, when shifting from the SI mode to the CI mode, the switching mode is executed before actually switching to the CI mode.

  The switching mode here performs control opposite to the switching mode when shifting from the CI mode to the SI mode described above. That is, if the load of the engine 1 decreases and the mode shifts from the SI mode to the switching mode, the valve closing timing of the intake valve 21 set on the retard side is gradually changed to the advance side, thereby As shown in FIG. 17, the filling amount is gradually increased. As described above, the throttle control of the throttle valve 36 may be performed instead of or together with the control of the intake valve 21. Further, the control of the EGR valve 511 may be used in combination.

  In the switching mode, the fuel injection amount is increased so as to satisfy λ = 1 in accordance with the increase in the filling amount (that is, fresh air). In addition, as will be described in detail later, the fuel pressure remains at a relatively high fuel pressure in the SI mode, and the generated torque is retarded more than the timing in the SI mode, respectively, by retarding the fuel injection timing and the ignition timing. Suppress. Also in this case, since the spark ignition combustion is performed by high pressure retarded injection with a relatively high fuel pressure, the combustion can be stabilized even if the ignition timing is greatly retarded.

  If the timing for shifting from the switching mode to the CI mode is reached, the closing timing of the intake valve 21 is set to the intake bottom dead center, while the exhaust valve 22 is opened twice through the control of the VVL 71. To do. As a result, the internal EGR gas is introduced into the cylinder 18 instead of the excessive fresh air introduced into the cylinder 18. Further, the fuel injection amount is reduced, and the fuel injection timing is changed to an intake stroke or a compression stroke. On the other hand, the operation of the spark plug 25 is substantially stopped, and a shift is made to the CI mode in which compression ignition combustion is performed.

  Thus, the switching mode related to the transition from the SI mode to the CI mode is a mode for increasing the filling amount in advance in preparation for the start of the internal EGR. Also in this case, the fuel injection amount is temporarily increased, and the sudden increase in the torque is suppressed by delaying the fuel injection timing and the ignition timing, and the filling amount is gradually increased to enter the CI mode. Move to smooth.

  Next, the control of the engine 1 will be described in more detail with reference to the flowchart shown in FIG. This flowchart is a control flow of the engine 1 executed by the PCM 10. By controlling the engine 1 in accordance with this flowchart, the state of the engine 1 with respect to load fluctuations (but steady) can be as shown in FIGS. This flow also includes control related to transition between the CI mode and the SI mode. The flow shown in FIG. 18 does not limit the execution order of the steps, and the order of the steps in the flow shown in FIG. 18 is an example. Therefore, it is possible to appropriately change the order of the steps or to execute a plurality of steps simultaneously in parallel. Moreover, it is also possible to omit a step suitably or to add another step with respect to the flow shown in FIG.

  First, in step SA1, the integrated AWS execution time is read, and in the subsequent step SA2, it is determined whether or not the read AWS execution time is equal to or greater than a predetermined value. The AWS (Accelerated Warm-up System) is a system that accelerates the purification of exhaust gas by increasing the temperature of the exhaust gas at the start of the engine 1 to accelerate the activation of the catalysts 41 and 42. The AWS is executed for a predetermined time after the engine 1 is started. Accordingly, when the determination at step SA2 is NO (that is, when the value is not equal to or greater than the predetermined value), the process proceeds to step SA3 to set the AWS mode. In the AWS mode, basically, SI combustion is executed in which the intake air amount is increased and the ignition timing of the spark plug 25 is largely retarded.

  On the other hand, when the determination in step SA2 is YES (that is, when the value is equal to or greater than a predetermined value), the process proceeds to step SA4. That is, AWS ends and the engine 1 shifts to a normal operation mode.

  In step SA4, the PCM 10 first reads the accelerator opening and the engine speed, and in the subsequent step SA5, the PCM 10 is detected by the intake air flow rate detected by the air flow sensor (AFS) SW1 and the in-cylinder pressure sensor (CPS) SW6. The filling amount is calculated based on the in-cylinder pressure. In step SA6, the PCM 10 reads the engine water temperature and the temperature of the intake air introduced into the cylinder 18, respectively. Then, the CI determination variable Y is calculated based on the accelerator opening, the engine speed, the filling amount, the engine water temperature, and the intake air temperature.

CI determination variable Y includes accelerator opening function i (accelerator opening), engine speed function j (1 / rpm), filling quantity function k (filling quantity), engine water temperature function l (engine water temperature). Based on each function of the intake air temperature function m (intake air temperature), for example, it is a variable calculated by the following equation.
CI decision variable Y
= I (accelerator opening) + j (1 / rpm) + k (filling amount) + l (engine water temperature) + m (intake air temperature)
The CI determination variable Y is a variable that serves as an index as to whether or not the air-fuel mixture undergoes compression ignition near the compression top dead center. In other words, the CI determination variable Y is a variable for determining whether the engine 1 should be operated in the CI mode or the SI mode. For example, as shown in FIG. 21A, when the CI determination variable Y is smaller than the first threshold value, there is a high possibility of misfire when attempting to operate in the CI mode. On the contrary, when the CI determination variable Y is equal to or greater than the second threshold value, there is a high possibility of pre-ignition when attempting to operate in the CI mode. It is possible to determine that it should be done. On the other hand, when the CI determination variable Y is equal to or greater than the first threshold value and less than the second threshold value, the air-fuel mixture is compressed and ignited at an appropriate timing near the compression top dead center. Can be determined.

  Returning to the flow of FIG. 18, in step SA7, the mode of the previous cycle is read, and in subsequent step SA8, based on the CI determination variable Y calculated in step SA6, it is determined whether or not the current cycle should be in the CI mode. To do. When the determination is YES, the process proceeds to step SA9, and when the determination is NO, the process proceeds to step SA12.

  In step SA9, it is determined whether or not the previous cycle is the CI mode. When the previous mode is the CI mode (that is, when the determination is YES), the process proceeds to step SA10 and the engine 1 is operated. The mode is changed to the CI mode (that is, the CI mode is continued). On the other hand, when the previous mode is the SI mode (that is, when the determination is NO), the process proceeds to step SA11 to set the operation mode of the engine 1 to the switching mode. This switching mode is a switching mode when switching from the SI mode to the CI mode.

  Similarly, in step SA12, it is determined whether or not the previous cycle is in the CI mode. When the previous mode is the SI mode (that is, when the determination is NO), the process proceeds to step SA13, and the operation mode of the engine 1 is set to the SI mode (that is, the SI mode is continued). On the other hand, when the previous mode is the CI mode, the process proceeds to step SA14 to set the operation mode of the engine 1 to the switching mode. This switching mode is a switching mode when switching from the CI mode to the SI mode.

  FIG. 19 shows a control flow relating to the CI mode in step SA10. First, in step SB1, the fuel pressure (target pressure) for the CI mode is read from a characteristic diagram set in advance and stored in the PCM 10. The characteristic diagram is set as a linear function with respect to the engine speed, as shown in FIG. 21B, and the target pressure is set to be higher as the engine speed is higher. The upper limit of the fuel pressure for the CI mode is a predetermined value (FP1).

  In the following step SB2, the high pressure fuel supply system 62 is controlled so that the fuel pressure becomes the target pressure, and in step SB3, the filling amount is controlled. As described with reference to FIGS. 11 to 14, the filling amount control includes at least exhaust double opening control by controlling the VVL 71, thereby introducing the internal EGR gas into the cylinder 18. Then, in step SB4, a separately set predetermined amount of fuel is directly injected into the cylinder 18 through the direct injection injector 67 at a predetermined timing in the intake stroke or the compression stroke.

  FIG. 20 shows a control flow relating to the SI mode in step SA13. First, in step SC1, the overall injection amount (which means the total amount of fuel injection injected per cycle) is read from a characteristic diagram that is preset and stored in the PCM 10. The characteristic diagram of the overall injection amount is set as a function of the accelerator opening, as shown in FIG. 21C, and is set so that the overall injection amount increases as the accelerator opening increases.

In the following step SC2, a knock variable X is calculated based on the engine speed, the intake pressure, the intake temperature, and the overall injection amount. The knock variable X includes an engine speed function a (1 / engine speed), an intake pressure function b (intake pressure), an intake temperature function c (intake temperature), and an overall injection amount function d (overall injection). Based on each function of (quantity), for example, it is a variable calculated by the following equation.
Knock variable X
= A (1 / engine speed) + b (intake air pressure) + c (intake air temperature) + d (overall injection amount)
The engine speed, the intake pressure, the intake temperature, and the overall injection amount are parameters related to the occurrence of knocking and pre-ignition, respectively, and the knock variable X is an index of the likelihood of abnormal combustion. That is, as the knock variable X is larger, abnormal combustion is more likely to occur, and conversely, as the knock variable X is smaller, abnormal combustion is less likely to occur. For example, the higher the engine speed, the smaller the knock variable X related to the reciprocal of the engine speed. In addition, the knock variable X increases as the overall injection amount increases, in other words, as the engine load increases.

  In step SC3, the fuel pressure (target pressure) for the SI mode is read from a characteristic diagram that is set in advance and stored in the PCM 10. As shown in an example of FIG. 21 (d), the characteristic diagram is different from the fuel pressure for CI combustion (FIG. 21 (b)) and is a function g (knock variable, engine speed) of the knock variable and the engine speed. ) As a linear function. For example, as the knock variable X increases, the target pressure is set higher. This is advantageous for avoiding abnormal combustion as described above. The lower limit value of the fuel pressure for the SI mode is set to a pressure (FP2) higher than the upper limit value FP1 of the fuel pressure for CI combustion. As a result, the fuel pressure for the SI mode is always higher than the fuel pressure for the CI mode. The fuel pressure on the high load side in the CI mode may be higher than the fuel pressure in the SI mode.

  In step SC4, the retard injection ratio R and the retard injection delay amount T are set based on the characteristic diagrams illustrated in FIGS. 21 (e) and 21 (f), respectively. The retard injection ratio R is a variable for setting a ratio between the fuel injection amount injected in the retard period and the intake stroke injection in the overall injection amount. The retard injection ratio R is set to be larger as the knock variable X is larger. Here, as will be described later, the fuel injection amount by the high pressure retarded injection is calculated by “total injection amount × retard injection ratio”, and the fuel injection amount by the intake stroke injection is “total injection amount × (1-retard injection ratio)”. ". Therefore, as the knock variable X is larger, the fuel injection amount by the high pressure retarded injection is increased, while the intake stroke injection is decreased. The retard injection ratio is a variable that is greater than 0 (zero) and 1 or less. When the retard injection ratio is 1, the total amount of the overall injection amount is injected by the high pressure retarded injection, and the intake stroke injection is not performed. Here, as shown in FIG. 21 (e), if the knock variable X is equal to or greater than a predetermined value, the retard injection ratio R is 1, so the intake stroke injection amount is 0, and intake stroke injection is not performed.

  Further, since the knock variable X decreases as the engine speed increases, the retard injection ratio becomes smaller than 1 in the medium speed range of the engine 1. As a result, as described above, the intake stroke injection is executed in the medium speed range of the engine 1 (see FIG. 10D).

  As shown in FIG. 21F, the retard injection delay amount T is set to be larger as the knock variable X is larger. In other words, the greater the knock variable X, the more retarded the injection timing of the high-pressure retarded injection. As described above, since the overall injection amount (engine load) and the knock variable X are proportional, the injection timing of the high pressure retarded injection is set to the retard side as the engine load increases. As the fuel injection timing is retarded, the ignition timing is also set to the retard side as the engine load increases. This is advantageous for avoiding abnormal combustion.

  In step SC5, the fuel injection amount is calculated from the following equations based on the read retard injection ratio.

Intake stroke injection amount = Overall injection amount × (1−retard injection ratio R)
High-pressure retarded injection amount = total injection amount × retarded injection ratio R
In step SC6, the ignition timing is read from an ignition map that is set in advance and stored in the PCM 10 as illustrated in FIG. This ignition map is a map for setting the ignition timing (IG) based on the engine speed and the accelerator opening. The lower the engine speed and the larger the accelerator opening, in other words, the upper the left of the map. The more the ignition timing is set to the retard side, the higher the engine speed and the smaller the accelerator opening, in other words, the closer to the lower right of the map, the more the ignition timing is set to the advance side ( IG1 <IG2 <IG3). The ignition timing set here is set to a timing later than the fuel injection timing described above.

  In this way, after setting the target fuel pressure, the fuel injection amount and fuel injection timing of the high pressure retarded injection, the fuel injection amount and fuel injection timing, and the ignition timing when performing the intake stroke injection, In step SC7, first, the high pressure fuel supply system 62 is controlled so that the fuel pressure becomes the target pressure, and in step SC8, the filling amount control is executed. As shown in FIGS. 11 to 14, the filling amount control is executed in order to set the air-fuel ratio λ = 1 in accordance with the set overall injection amount in the SI mode operated at the air-fuel ratio λ = 1. The control for restricting the intake air introduced into the cylinder 18, the control for introducing the external EGR gas into the cylinder 18, or the combination of both is executed.

  In step SC9, the intake stroke injection of the set fuel injection amount is executed at the set injection timing. In this step, as described above, the fuel is injected into the intake port 16 by the port injector 68. However, when the fuel injection amount of the intake stroke injection is set to 0, step SC9 is substantially omitted.

  In step SC10, high pressure retarded injection of the set fuel injection amount is executed at the set injection timing. Therefore, this injection timing is within the retard period from the latter stage of the compression stroke to the early stage of the expansion stroke, and the fuel is directly injected into the cylinder 18 by the direct injection injector 67. Note that, as described above, this high-pressure retarded injection may be divided injection including two fuel injections of the first injection and the second injection executed within the retard period, for example, depending on the fuel injection amount. In step SC11, ignition by the spark plug 25 is performed at the set ignition timing.

  FIG. 22 shows a control flow relating to the switching mode of step SA11 or SA14. The control flow is the same when switching from the CI mode to the SI mode and when switching from the SI mode to the CI mode, but the characteristic diagrams used in each step are from the CI mode to the SI mode. And when switching from the SI mode to the CI mode. Here, first, the control flow will be described by taking the switching mode when switching from the CI mode to the SI mode as an example.

  First, in step SD1, the fuel injection amount is read from a characteristic diagram that is preset and stored in the PCM 10. The characteristic diagram is set as a linear function over time as shown in FIG. 23A. That is, when the CI mode is switched to the switching mode, the fuel injection amount is greatly increased. This fuel injection amount is larger than the fuel injection amount corresponding to the load of the engine 1 and is substantially the same as the fuel injection amount at the fully open load. The fuel injection amount is gradually decreased with the passage of time, and is set to be a predetermined fuel injection amount at the end of the switching mode (in other words, at the time of transition from the switching mode to the SI mode). This is, for example, a fuel injection amount corresponding to the load of the engine 1. Although omitted in the flow of FIG. 22, the fuel pressure is set as the fuel pressure for the SI mode according to step SC3 in the SI mode.

  In the subsequent step SD2, the valve closing timing (IVC) of the intake valve 21 is read from the characteristic chart set in advance and stored in the PCM 10. The characteristic diagram is set as a linear function with respect to the passage of time as shown in FIG. 23 (b) as an example. From the relatively advanced valve closing timing in the CI mode, The valve timing is gradually changed with the passage of time to the valve closing timing on the retard side. By advancing the valve closing timing of the intake valve 21 in this way, as shown in FIG. 17, the intake charge amount (fresh air amount) gradually decreases in the switching mode. Therefore, as set in step SD1, the stoichiometric air-fuel ratio λ = 1 is maintained in the switching mode in combination with the gradually decreasing fuel injection amount.

  In step SD3, the fuel injection timing is read from a characteristic diagram that is preset and stored in the PCM 10. The characteristic diagram is set as a linear function with respect to the passage of time as shown in an example in FIG. 23 (c), and the injection timing is relatively advanced during the transition from the CI mode to the switching mode. It is greatly changed from the timing on the corner side to the retard side. Thereafter, it is gradually advanced with the passage of time, and is changed to the fuel injection timing in the SI mode. Therefore, in the switching mode, the fuel injection timing is delayed from the fuel injection timing in the SI mode (that is, the fuel injection timing within the retard period). As described above, when the split injection is performed in the SI mode, the fuel injection timing within the retard period, which is the last fuel injection among the plurality of fuel injections, may be retarded.

  In step SD4, the ignition timing is read from a characteristic diagram that is preset and stored in the PCM 10. The characteristic diagram is set to have the same characteristics as the characteristic diagram of the fuel injection timing, as shown in FIG. 23 (d) as an example. That is, at the time of transition from the CI mode to the switching mode, the ignition timing is set to a substantially retarded side as compared with the ignition timing in the SI mode, and is gradually advanced as time elapses. To the ignition timing of the hour. The generation of torque is suppressed by the retard control of the fuel injection timing and the ignition timing by the steps SD3 and SD4. As a result, torque shock in the switching mode is avoided, and the mode transitions smoothly.

  Then, in step SD5, fuel injection and ignition are executed with a relatively high fuel pressure in accordance with the fuel injection amount, fuel injection timing and ignition timing set in the previous steps, and in step SD6, which is preset in step SD6. It is determined whether the predetermined time has elapsed. This predetermined time corresponds to the time for executing the switching mode, and when this determination is NO (when the predetermined time has not elapsed), the process returns to step SD1 to continue the switching mode. On the other hand, when this determination is YES (when a predetermined time has elapsed), the process proceeds to step SD7, and the flow is terminated assuming that the mode switching is completed.

  Next, the switching mode at the time of switching from the SI mode to the CI mode will be described with reference to the characteristic diagrams of FIGS. Each of these characteristic diagrams is basically opposite to the above-described characteristic diagrams of the switching mode when switching from the CI mode to the SI mode.

  First, FIG. 23E is a characteristic diagram of the fuel injection amount, and the fuel injection amount is set so as to gradually increase with the passage of time. Thus, when shifting from the switching mode to the CI mode, the fuel injection amount corresponding to the engine load in the CI mode is significantly reduced. As described above, the fuel pressure is set as the fuel pressure for the SI mode in accordance with step SC3 in the SI mode.

  FIG. 23 (f) is a characteristic diagram of the closing timing (IVC) of the intake valve 21, and from the relatively retarded valve closing timing in the SI mode to the relatively advanced angle side in the CI mode. The valve closing timing is continuously changed over time. By advancing the valve closing timing of the intake valve 21 as described above, the filling amount gradually increases in the switching mode as shown in FIG.

  FIG. 23 (g) is a characteristic diagram of fuel injection timing. From the relatively retarded injection timing in the SI mode (that is, the fuel injection timing within the retard period), the timing is further retarded over time. Thus, at the time of transition from the switching mode to the CI mode, the fuel injection timing is significantly advanced to the relatively advanced timing. As described above, when the split injection is performed in the SI mode, the fuel injection timing within the retard period, which is the last fuel injection among the plurality of fuel injections, may be retarded.

  FIG. 23 (h) is a characteristic diagram of the ignition timing. Similar to the characteristic diagram of the fuel injection timing, the ignition timing is gradually retarded with the passage of time from the ignition timing in the SI mode.

  Thus, even in the switching mode related to the transition from the SI mode to the CI mode, the generation of torque is suppressed by retarding the fuel injection timing and the ignition timing, and as a result, the torque shock is avoided. In addition, since the fuel pressure is set to be relatively high, combustion can be stabilized.

  In the above configuration, for example, as clearly shown in FIG. 11 and the like, the switching between the CI mode and the SI mode coincides with the execution and cancellation of the double opening control of the exhaust valve 22. In other words, the internal EGR gas is always introduced into the cylinder 18 when the operation state of the engine 1 is in the low load to medium load range in the low speed range and the compression ignition combustion is performed. On the other hand, when the operating state of the engine 1 is in the middle load range in the low speed range, the CI valve mode may be performed while stopping the internal EGR by stopping the opening control of the exhaust valve 22 twice. . That is, when the operating region of the engine 1 is in the middle load region in the low speed region, the compression ignition combustion may be performed without introducing the internal EGR gas into the cylinder 18.

  Further, the port injector 68 and the low pressure fuel supply system 66 may be omitted, and the intake stroke injection may be performed by the direct injection injector 67.

1 Engine (Engine body)
10 PCM (controller)
18 cylinder 25 spark plug 62 high pressure fuel supply system (variable fuel pressure)
67 Direct injection injector (fuel injection valve)

Claims (8)

An engine body having a cylinder with a geometric compression ratio set to 14 or more and configured to be supplied with fuel containing at least gasoline;
A fuel injection valve configured to inject the fuel into the cylinder;
An ignition plug disposed facing the cylinder and configured to ignite an air-fuel mixture in the cylinder;
A variable fuel pressure mechanism configured to change the pressure of fuel injected by the fuel injection valve;
A controller configured to operate the engine body by controlling at least the fuel injection valve, the spark plug, and the fuel pressure variable mechanism; and
The controller is
When the operating state of the engine body is in a predetermined low load region, a compression ignition mode for executing compression ignition combustion is set,
In the high load region where the load is higher than that in the compression ignition mode, the fuel pressure variable mechanism is controlled so that the fuel pressure is higher than at least in the low load region in the compression ignition mode, and the expansion stroke is started from the late stage of the compression stroke. Spark ignition that drives the fuel injection valve so as to include at least a fuel injection performed at a specific timing within the retard period until and drives the spark plug so as to ignite after the fuel is injected within the retard period Mode and
The controller also includes a fuel pressure in the spark ignition mode within a predetermined transition period when switching between the compression ignition mode and the spark ignition mode in accordance with a change in the load of the engine body, and A control apparatus for a spark ignition gasoline engine, wherein the fuel injection valve is driven so as to inject fuel at a timing retarded from the specific timing, and the ignition plug is driven after the fuel injection. In the control device of the spark ignition type gasoline engine according to claim 1,
In the spark ignition mode, the controller operates the engine body at an air-fuel ratio λ = 1, and in the compression ignition mode, the controller operates the engine body at a leaner air-fuel ratio λ = 1. Control device for gasoline engine. In the control device of the spark ignition type gasoline engine according to claim 1 or 2,
In the switching mode, the controller is a control device for a spark ignition gasoline engine that increases the fuel injection amount more than a fuel injection amount corresponding to an engine load. In the control device of the spark ignition type gasoline engine according to claim 3,
In the compression ignition mode, the controller performs internal EGR control in which a part of burned gas exists in the cylinder, and in the switching mode, the air-fuel ratio is λ == without performing the internal EGR control. A spark-ignition gasoline engine control device that performs air-fuel ratio control to be 1. In the control device of the spark ignition type gasoline engine according to any one of claims 1 to 4,
The fuel pressure variable mechanism is a control device for a spark ignition gasoline engine that sets the fuel pressure to 40 MPa or more in the spark ignition mode and the switching mode. In the control device of the spark ignition type gasoline engine according to any one of claims 1 to 5,
The fuel injection valve has a plurality of injection holes, and is a control device for a spark ignition gasoline engine configured so that fuel spray injected from these injection holes spreads radially in the cylinder. In the control device of the spark ignition type gasoline engine according to any one of claims 1 to 6,
In the spark ignition mode, the controller drives the fuel injection valve to execute a plurality of times of fuel injection within the retard period, and in the switching mode, of the plurality of times of fuel injection. A control device for a spark ignition gasoline engine that retards the timing of the last fuel injection. In the control device of the spark ignition type gasoline engine according to any one of claims 1 to 7,
In the compression ignition mode, the controller drives the fuel injection valve so as to inject fuel at a timing more advanced than the fuel injection timing in the retard period in the spark ignition mode, and A control device for a spark ignition gasoline engine that compresses and ignites an air-fuel mixture in the cylinder by causing a part of burned gas of the engine body to exist in the cylinder.

Images