JPH11141386A - Method and device for controlling pilot fuel injection quantity for engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve stable operation of an engine and exhaust gas performance by a method wherein a pilot injection quantity is controlled so that a peak value of a heat generating factor along with piot injection calculated based on a pressure in a combustion chamberm concides with a target peak value. SOLUTION: A peak value qmax of a heat generating factor q is calculated as a maximum value of a heat generating factor during a calculating period between crank angles θs and θe. Based on a deviation between the peak value qmax of the calculating pilot heat generating factor q and a pilot heat generating factor q predetermined according to the operation stage of an engine, the pulse width PWp of a drive pulse for pilot infection. Excellent combustion of infection fuel by pilot infection is effected, the generation of engine noise is suppressed and exhaust gas performance is improved.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for controlling a pilot combustion injection amount of an engine in which a pilot injection is injected into a combustion chamber prior to a main injection.

[0002]

2. Description of the Related Art In a diesel engine, as a method of controlling fuel injection, a fuel injection pressure is increased and an injection characteristic such as a fuel injection timing and an injection amount is optimally controlled according to an engine operating state. A common rail type fuel injection system is known. A common rail fuel injection system stores fuel pressurized to a predetermined pressure by a fuel pump in a common rail common to a plurality of injectors, and controls the fuel stored in the common rail from each injector under control of a controller. This is a fuel injection system for injecting indoors. The controller controls the fuel pressure of the common rail and the operation of the control valve provided in each injector so that the fuel is injected in each injector under optimal injection conditions for the operating state of the engine. In order to optimally control the operation of the engine, the common rail fuel injection system maps a predetermined fuel injection amount and a predetermined fuel injection timing in accordance with the operation state of the engine and stores them in the controller. From the obtained fuel injection amount and fuel injection timing, a target fuel injection amount and target fuel injection timing corresponding to the current operating state of the engine are obtained, and the obtained target fuel injection amount and target fuel injection timing are compared with the current fuel injection amount and The fuel injection valve of the injector is electronically controlled based on the deviation from the fuel injection timing, and fuel is injected from an injection hole formed in the injector.

[0003] The structure of the injector 3 will be briefly described with reference to FIG. FIG. 21 is a longitudinal sectional view of the injector. The injector 3 is mounted in a sealed state by a seal member in a hole provided in a base such as a cylinder head, but the structure of the cylinder head and the like is not shown. A branch pipe 23 branched from a common rail is connected to an upper side of the injector 3 via a fuel inlet joint 60. Fuel passages 61 and 62 are formed inside the main body of the injector 3, and a fuel passage is formed by the branch pipe 23 and the fuel passages 61 and 62. The fuel supplied through the fuel flow path is injected into the combustion chamber through a fuel reservoir 63 and a passage around the needle valve 64 from an injection hole 65 opened when the needle valve 64 is lifted.

[0004] The injector 3 is provided with a balance chamber type needle valve lift mechanism for controlling the lift of the needle valve 64. That is, an electromagnetic actuator 66 is provided at the uppermost part of the injector 3, and a control current as a control signal from the controller 37 is sent to an electromagnetic solenoid 68 of the electromagnetic actuator 66 through a signal line 67. When the electromagnetic solenoid 68 is excited,
Since the armature 69 rises and opens the on-off valve 72 provided at the end of the fuel passage 71, the fuel pressure of the fuel supplied from the fuel passage to the balance chamber 70 is released through the fuel passage 71. A control piston 74 is provided in a hollow hole 73 formed inside the main body of the injector 3 so as to be able to move up and down. Reduced balance chamber 70
The force pushing up the control piston 74 based on the fuel pressure acting on the tapered surface 76 facing the fuel reservoir 63 is superior to the pushing force acting on the control piston 74 due to the force based on the internal pressure and the spring force of the return spring 75. Therefore, the control piston 74 moves up.
As a result, lift of the needle valve 64 is allowed, and fuel is injected from the injection hole 65. The fuel injection amount is determined by the fuel pressure in the fuel passage and the lift (lift amount, lift period) of the needle valve. The lift of the needle valve 64 is determined by a drive pulse as a control current sent to the electromagnetic solenoid 68 to control the opening and closing of the on-off valve 72. The needle valve 64
Of the valve shaft 77 can slide through a small gap in the hollow hole 79 of the nozzle 78, and the tapered portion 80 formed at the tip of the valve shaft 77 contacts the corresponding tapered portion of the nozzle 78 to jet. The hole 65 can be opened and closed.

In general, FIG. 22 shows the relationship between the fuel injection amount Q of the injector 3 and the injection pulse width W supplied from the controller 37 to the electromagnetic solenoid 68, based on the fuel pressure Pcr.
(Fuel pressure in the common rail 2) is shown as a parameter. Assuming that the fuel pressure Pcr is constant, the fuel injection amount Q increases as the injection pulse width W increases, and even with the same injection pulse width W, the fuel injection amount Q increases as the fuel pressure Pcr increases. . On the other hand, since fuel injection is started or stopped with a certain time delay from the rise time and fall time of the injection pulse, it is necessary to control the injection timing by controlling when the injection pulse is turned on or off. it can.

[0006] Incidentally, a diesel engine is liable to generate combustion noise when it is in a low-speed, low-load operation state such as an idling operation. Such combustion noise is generated due to fuel ignition delay. Therefore, it is known that it is effective to inject a part of the total fuel injection amount in the combustion cycle by pilot injection (preliminary injection) prior to main injection as a means to cope with combustion noise. Have been. The fuel injected by the pilot injection is burned to sufficiently raise the temperature of the wall of the combustion chamber, and then the remaining main fuel amount is injected (main injection). Combustion, that is, ignition delay of the main injection can be avoided. Further, since the fuel injection is divided into the pilot injection and the main injection, the initial combustion is suppressed, and the NOx can be reduced. That is, a certain amount of air-fuel mixture is burned in the early stage of combustion, and the amount of air-fuel mixture burned by the main injection is reduced, so that the temperature of the combustion gas can be suppressed and the generation amount of NOx decreases.

However, the fuel injection characteristics of the injectors 3 provided in the respective cylinders vary depending on various factors such as processing errors and assembly errors of the injectors 3 and the length of the branch pipe 23 connecting the common rail and the injectors 3. Therefore, the amount of fuel actually injected by each injector 3 (hereinafter referred to as the actual fuel injection amount) varies, making it difficult to match the actual fuel injection amount with the target fuel injection amount.

As described above, the pilot injection is an effective means for preventing the combustion noise and the generation of NOx. However, in the idle operation state where the fuel injection amount is extremely small, as shown in FIG. It becomes difficult to control the fuel injection amount Q by the magnitude of the pressure Pc and the injection pulse width W. That is, in the region where the fuel injection amount is large, the fuel injection amount can be effectively controlled by changing the common rail pressure Pcr as a parameter and by changing the injection pulse width W supplied to the electromagnetic actuator of the injector 3. In the region where the fuel injection amount is small, it is difficult to finely control the fuel injection amount for both the common rail pressure Pcr and the injection pulse width W.

In this situation, as shown in FIG. 23, when the target fuel injection amount is a large value (Q1), the target fuel injection amount Q1 in the fuel injection amount variation range ΔQ1 even if the injection pulse width is the same. Is relatively small, but in the case of pilot injection, since the fuel injection amount Q2 is a small value, the variation range of the fuel injection amount ΔQ2 for the same pilot injection pulse width is obtained. It can also be understood from the fact that the ratio to the target fuel injection amount Q2 becomes very large.

Therefore, originally, in the case of the pilot injection having a small fuel injection amount, it may not be able to play the role of the pilot injection due to the variation of the fuel injection characteristics of each injector 3. Further, the fuel flow characteristics from the common rail to each injector 3 may change with time, and the role of the pilot injection may be initially expected but may not be fulfilled with the passage of time. Since the pilot injection has the above phenomena, the pilot injection may not be performed in some cylinders even when the pilot injection is performed, and the combustion noise and the NOx reduction effect of the pilot injection are reduced. There is a problem.

By the way, as a combustion injection control device for a diesel engine, the start timing of main injection is determined by detecting the peak of combustion of fuel injected by pilot injection and its vicinity.
Japanese Patent Application Laid-Open No. 2-95751 discloses a method in which one or both of the main injection timing and the pilot injection timing is controlled so as to be at or near the combustion peak of fuel by pilot injection. The detection of the peak of the combustion of the fuel injected by the pilot injection or the main injection is performed by detecting the pressure waveform in the combustion chamber, replacing the pressure waveform with the heat generation rate, and calculating the heat generation peak corresponding to the pilot injection or the main injection from the heat generation pattern. This is done by detecting.

Japanese Patent Application Laid-Open No. 62-17343 discloses a technique for controlling the pilot injection timing for the purpose of reducing the combustion noise of the engine. Attach a sensor to detect noise level near the engine,
An increase or decrease in the noise level is detected by slightly advancing or retreating the pilot injection timing, and the pilot injection timing is feedback-controlled in a direction that minimizes engine noise.

[0013]

Therefore, in the engine, it is detected whether or not a required amount of fuel has been injected by the pilot injection, and based on the detection result, the fuel injection characteristics of the individual injectors vary or vary with time. Even if there is a change, there is a problem that it is not possible to secure a necessary pilot injection amount in the pilot injection.

[0014]

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned problem. In an engine in which fuel injected from an injector ignites and burns, fuel injected by pilot injection and main injection is used. Focusing on the fact that by detecting the pressure in the combustion chamber that changes when the fuel is ignited and calculating the heat release rate based on the pressure in the combustion chamber, it is possible to know whether the fuel has been sufficiently ignited and burned by pilot injection Then, based on the calculation result, the actuator of the injector is driven so that the heat generation rate (hereinafter, referred to as pilot heat generation rate) appearing when the fuel injected by the pilot injection ignites and burns becomes a predetermined level. By correcting the pilot injection pulse width, the fuel injection characteristics of individual injectors vary. Even if there is aging, it is to provide a pilot injection amount control method and apparatus that can carry out the injection of the required amount of fuel in the pilot injection.

According to the present invention, the fuel injected from the injector into the combustion chamber is divided and injected into a main injection and a pilot injection preceding the main injection, and the fuel is injected based on the sequentially detected pressure in the combustion chamber. A heat release rate in the room is calculated, and a peak value of a pilot heat release rate caused by the ignition of the fuel injected in the pilot injection is calculated from the heat release rate, and is determined in advance based on an operation state of the engine. The target peak value of the pilot heat release rate according to the current operating state of the engine is obtained from the target peak value of the pilot heat release rate, and the peak value of the pilot heat release rate and the target peak value are obtained. Correcting the pilot fuel injection amount injected by the pilot injection so as to eliminate the deviation from the value. For the pilot fuel injection amount control method in the engine.

Further, the present invention provides an injector for injecting the fuel supplied through a fuel passage into a combustion chamber of an engine, and injects the fuel into a main injection and a pilot injection preceding the main injection. A controller that controls the driving of the injector, calculates the rate of heat generation in the combustion chamber based on the sequentially detected pressure in the combustion chamber, and performs injection in the pilot injection based on the rate of heat generation. Calculating a peak value of the pilot heat release rate caused by the ignition of the fuel, and calculating a current peak value of the engine from the target peak value of the pilot heat release rate determined in advance based on an operation state of the engine. Determining the target peak value of the pilot heat release rate according to the operating state;
The present invention relates to a pilot fuel injection amount control device for an engine, which corrects a pilot fuel injection amount injected by the pilot injection so as to eliminate a deviation between the peak value of the pilot heat release rate and the target peak value.

According to the method and apparatus for controlling the amount of pilot fuel injection in the engine according to the present invention, the rate of heat generation in the combustion chamber is calculated based on the sequentially detected pressure in the combustion chamber. The peak value of the pilot heat release rate generated due to the ignition of the fuel that has been generated is calculated. On the other hand, the target peak value of the pilot heat release rate according to the current operation state of the engine is obtained from the target peak value of the pilot heat release rate which is predetermined based on the operating state of the engine. By correcting the pilot fuel injection amount injected by the pilot injection, the peak value of the pilot heat generation rate is controlled so as to match the target peak value, so that the fuel injection characteristics of the individual injectors vary and change over time. Even if there is, the pilot injection amount required in the pilot injection is always secured, and the operation stability of the engine, the exhaust gas performance and the like can be maintained satisfactorily.

In the method and apparatus for controlling the amount of pilot fuel injection in an engine according to the present invention, the volume in the combustion chamber is determined sequentially, and the heat release rate in the combustion chamber is determined based on the sequentially detected pressure in the combustion chamber and the sequential determination. It is calculated from the determined volumes in the combustion chamber and their rate of change.

In the method and apparatus for controlling the amount of pilot fuel injection in an engine according to the present invention, the peak value of the pilot heat release rate is determined by the pilot injection driving pulse supplied to the injector to perform the pilot injection. The calculation period between the calculation start time after the start delay has elapsed from the start timing and the calculation end time after the end delay has elapsed from the main injection timing at which the main injection drive pulse is supplied to the injector to perform the main injection. It is obtained as the maximum value of the heat release rate within the calculation period.

In the method and apparatus for controlling the amount of pilot fuel injection in an engine according to the present invention, the start delay is defined as the injection delay of the injector from the pilot injection timing until the injector actually starts the pilot injection and the pilot injection from the injector. This is the sum of the ignition delay until the injected fuel is ignited, and the end delay is the injection delay of the injector from the main injection timing until the injector actually starts the main injection and the injection from the injector as the main injection. This is a delay obtained by adding the ignition delay until the fuel ignited.

Further, in the method and apparatus for controlling a pilot fuel injection amount in an engine according to the present invention, the fuel supplied to the injector is lifted by a needle valve in response to the driving of an electromagnetic actuator provided in the injector. This is performed by opening the injection hole formed at the tip, and the pilot fuel injection amount is corrected by changing the drive pulse width to the electromagnetic actuator of the injector. For example, the correction amount of the drive pulse width of the injector to the electromagnetic actuator is:
It is obtained by correcting the correction amount of the previous drive pulse width based on the difference between the maximum heat release rate and the target maximum heat release rate. Further, the drive pulse width to the electromagnetic actuator, which is calculated based on the pilot injection amount obtained according to the operating state of the engine, is corrected by the correction amount of the drive pulse width this time. It is set as the final drive pulse width.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A pilot injection amount control method and apparatus according to the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing an embodiment of an engine to which a pilot injection amount control device according to the present invention is applied, and FIG. 2 is a graph showing changes in in-cylinder pressure, fuel injection rate, and heat generation rate with the passage of crank angle. FIG. 3 is a graph showing the outline of the in-cylinder pressure and data processing in each injector as the crank angle elapses. In addition, what is already known, such as the structure of the injector 3, can be adopted as an embodiment of the present invention as long as it does not contradict the present invention, and will be appropriately used in the description of the embodiment.

FIG. 1 shows an outline of a common rail fuel injection system to which a method and a device for controlling pilot injection of an engine according to the present invention are applied. In this common rail fuel injection system, the engine is a four-cylinder engine. Injectors 31, which inject fuel into a combustion chamber (not shown) formed in the cylinder 2,
Fuel is supplied to 32, 33, and 34 (3 is used when collectively referred to) from the common rail 22 through a branch pipe 23 constituting a part of a fuel flow path. Feed pump 2
6 pressurizes the fuel sucked through the filter 25 from the fuel tank 24 to a predetermined pressure, and sends the pressurized fuel to the fuel pump 20 through the fuel pipe 27. The fuel pump 20 is a so-called plunger type supply fuel supply pump which is driven by, for example, an engine, boosts fuel to a high pressure determined based on the operating state of the engine, and supplies the fuel to the common rail 22 through a fuel pipe 29. is there. The fuel is raised to a predetermined pressure and the common rail 22
And is supplied to each injector 3 from the common rail 22. The injector 3 injects fuel into a corresponding combustion chamber at an appropriate injection timing and injection amount under the control of a controller 37 which is an electronic control unit. The injection pressure of the fuel injected from the injector 3 is
Since the pressure of the fuel stored in the fuel tank 2 is substantially equal to the pressure of the common rail, that is, the common rail pressure, the common rail pressure is controlled to control the injection pressure. The fuel relieved from the fuel pump 20 is returned to the fuel tank 24 through the return pipe 35. Further, of the fuel supplied from the branch pipe 23 to the injector 3, the fuel not consumed for injection into the combustion chamber is returned to the fuel tank 24 through the return pipe 36.

The controller 37 includes a crank angle sensor 10 for detecting the engine speed Ne and an accelerator opening sensor 1 for detecting the accelerator opening Acc.
1, a common rail pressure sensor 12 provided on the common rail 22 for detecting a fuel pressure (common rail pressure) in the common rail 22, and a water temperature sensor 13 for detecting a cooling water temperature. Signal from sensor is input. In addition to these sensors, for example, an intake pipe pressure sensor for detecting the intake pipe pressure can be included. Based on these signals, the controller 37 controls the fuel injection characteristics of each injector 3, that is, the fuel injection timing and injection amount, so that the engine output becomes the optimum output according to the operating state. Even if the fuel in the common rail 22 is consumed by the injection of the fuel from the injector 3,
The controller 37 controls the discharge amount of the fuel pump 20 so that the common rail pressure Pcr becomes constant.

The multi-cylinder engine 1 shown in FIG. 1 is a four-cylinder engine. The four cylinders # 1 to # 4 are respectively provided with injectors 31, 32, 33, and 34 (in general, 3 is used) for injecting fuel into a combustion chamber (not shown) formed in the cylinder 2. In-cylinder pressure sensor 4 as pressure detecting means for detecting the pressure in the combustion chamber (in-cylinder pressure)
1, 42, 43, and 44 (when collectively referred to, 4 is used). The in-cylinder pressure sensor 4 is connected to each cylinder # 1
It is installed facing the # 4 combustion chamber. A signal indicating the in-cylinder pressure of each of the cylinders # 1 to # 4 detected by the in-cylinder pressure sensor 4 is input to the controller 37. Further, a sensor provided in the multi-cylinder engine 1 to obtain information on the rotation of the engine, that is, a cylinder discriminating sensor 8 for discriminating a reference cylinder, a sensor B for detecting a crank angle position before a compression top dead center.
TDC (before topdead center)
r) The signals detected by the sensor 9 and the crank angle sensor 10 are also input to the controller 37.

FIG. 2 is a graph showing changes in the in-cylinder pressure (combustion chamber pressure) Pc, the fuel injection rate Rf, and the heat release rate q with respect to the crank angle. The change in the in-cylinder pressure Pc when the fuel injection is not performed changes symmetrically with respect to the peak of the in-cylinder pressure Pc with the change in the crank angle, but the fuel injection is performed at time T0. In addition, when the injected fuel ignites, the rise in the in-cylinder pressure Pc becomes slightly gentle, and then increases greatly. It is difficult to accurately determine the crank angle as the fuel ignition timing from the rapidly increasing in-cylinder pressure Pc. The fuel injection rate Rf shown in the middle graph is controlled so as to have two peaks, an initial injection and a main injection.

FIG. 3 is a graph showing the outline of the change of the in-cylinder pressure and the generation of various signals with the passage of the crank angle. Since the engine is a 4-cycle engine, the count value θ of the crank angle which counts 1 every 1 ° of the crank angle is 0 at the compression top dead center of the # 1 cylinder, and the crankshaft rotates twice, that is, the count value. One round at 719. FIG.
The four graphs in the upper row show the change in the in-cylinder pressure Pc with the passage of the count value θ of the crank angle of each of the cylinders # 1 to # 4. Since the engine 1 has four cylinders, assuming that the cylinder numbers n are # 1 to # 4 according to the arrangement of the columns, the combustion order i is # 1 → # 3 → # 4 → # 2 as shown in Table 1.
It becomes the order of.

[Table 1] In each of the cylinders # 1 to # 4, the compression and explosion strokes are successively performed in the above order, and combustion is performed. When a certain cylinder passes through the explosion stroke, the next cylinder enters the compression stroke. The output of each sensor regarding the rotation of the engine is shown in the middle graph of FIG. Since the cylinder # 1 is a reference cylinder, a cylinder discrimination (REF) signal is output at 120 ° before the compression top dead center. Also, a BTDC signal is output at 60 ° before the compression top dead center of each cylinder. For cylinder # 1, 1 before compression top dead center
The cylinder # 1 undergoes the compression / explosion stroke from 80 ° to 180 ° after the top dead center of compression, that is, when the count value of the crank angle θ is 540 or more and less than 180. The internal pressure Pc is detected, and the detection data is stored. The main processing is calculated within a predetermined time from 180 ° after the compression top dead center based on the stored data, and the processing of the injector 31 is performed based on the next BTDC interrupt signal. The lower graph in FIG. 3 schematically shows the processing order and timing of each injector 3 in each of the cylinders # 1 to # 4 as the count value θ of the crank angle increases.

FIG. 4 shows the relationship between various sensors centered on the controller 37 of the engine and the injector 3. FIG. 4 shows a controller 3 for performing fuel injection control including pilot injection amount control of a multi-cylinder engine that receives detection signals from various sensors related to engine rotation and a cylinder pressure sensor and outputs a control signal to each injector.
FIG. 7 is a block diagram of FIG. The rotation sensor of the engine 1 includes a reference cylinder among the cylinders # 1 to # 4, for example, a cylinder discrimination (REF) sensor 8 for discriminating the position of 120 ° before the top dead center of # 1, and the cylinders # 1 to # 4. 60 ° before top dead center of the explosion process of 4
BTDC (before top dead)
(center) sensor 9 and a crank angle sensor 10 for detecting a crank angle every 1 °. Since the engine 1 is a four-cycle engine, the cylinder pump sensor 20 and the BTDC sensor 9 generate four BTDC signals and one REF signal during two rotations of the crankshaft, respectively. And a cam shaft for driving the intake and exhaust valves. The cylinder discriminating signal detected by the cylinder discriminating REF sensor and the crank angle signal detected by the crank angle sensor are input to the CPU 14 in parallel with a DSP (digital sig).
nal processor 15 is also input. D
The SP 15 can add / subtract the input signal at high speed.

In the controller 37, in addition to the sensors 8 to 10 relating to the rotation of the engine 1, a common rail pressure sensor for detecting the pressure of the accelerator opening sensor 11, a common rail and the like is used to indicate the operating state of the engine 1. 12 and a detection signal from a water temperature sensor 13 for detecting the temperature of cooling water for cooling the engine 1 or an intake pressure sensor are input to a central processing unit (CPU) 14.

The exchange of data between the CPU 14 and the DSP 15 is performed via a dual port memory 16 which is a common RAM readable and writable from both sides of the CPU 14 and the DSP 15. The CPU 14 and the dual port memory 16 are connected through a CPU bus 17, and the DSP 15 and the dual port memory 16 are connected through a DSP bus 18. In-cylinder pressure P
The in-cylinder pressure sensors 4 for detecting c are in-cylinder pressure sensors 41 to 4 provided facing the combustion chambers of the cylinders # 1 to # 4.
4, the in-cylinder pressure Pc is detected as a relative pressure (gauge pressure) with respect to the atmospheric pressure. Analog signals of the in-cylinder pressure detected by the in-cylinder pressure sensors 41 to 44 are input to the AD converter 19, converted into digital signals, and sent to the DSP 15 through the DSP bus 18.

The CPU 14 controls the DSP 1 with respect to the information indicating the operating state of the engine 1 which is directly input from the sensors 8 to 13 and the in-cylinder pressure from the in-cylinder pressure sensors 41 to 44.
5 to calculate the fuel injection timing and fuel injection amount of the injectors 31 to 34 provided corresponding to each of the cylinders # 1 to # 4. Do. The DSP 15 performs high-speed addition and subtraction processing of digital signals related to the in-cylinder pressure Pc. Since this process is a digital process, the differentiation and integration of the in-cylinder pressure Pc can be similarly calculated at a high speed. Also, CP
U15 controls the discharge amount of the variable fuel pump 20 to control the pressure of the common rail, and controls the EGR valve 21 to control the exhaust gas circulation amount.

The fuel injection control by the CPU 14 is based on the operating state of the multi-cylinder engine 1, that is, the accelerator opening sensor 11.
A fuel injection amount characteristic map predetermined based on signals from the sensors 8 to 10 relating to the rotation of the engine 1 is stored, and a target fuel injection amount according to the current operating state of the engine is obtained from the characteristic map. Can be The pilot injection amount is obtained from the target fuel injection amount and the engine speed. In the pilot injection amount control according to the present invention, ignition of fuel injected by pilot injection
The peak value of the pilot heat release rate based on the combustion is obtained from the in-cylinder pressure Pc detected by the in-cylinder pressure sensors 41 to 44, and the obtained peak value of the heat release rate is the peak value of the target heat release rate based on the pilot injection. The pilot injection pulse width supplied to the electromagnetic actuator of the injector is corrected so as to coincide with. Hereinafter, the control of the pilot injection amount will be described. In this description, the fuel injection timing of each injector 3 is corrected in consideration of the ignition delay so that the actual combustion ignition timing obtained from the in-cylinder pressure detected by each in-cylinder pressure sensor 4 matches the target combustion ignition timing. This is performed together with the correction of the fuel injection timing to be performed.

The CPU 14 performs the main processing shown in FIG. FIG. 5 shows the CPU in the controller shown in FIG.
5 is a flowchart showing main processing of FIG. This main processing includes the following steps. (1) Initialization of the CPU 14 is performed (steps 1, S
Abbreviated as 1. same as below). (2) Process the sensor signal (S2). As shown in FIG. 4, processing of detection signals from various sensors input to the CPU 14 is performed. (3) Based on the information obtained in the signal processing performed in S2, the amount of fuel to be injected by each injector 3, that is, the fuel injection amount is calculated (S3). The calculation of the fuel injection amount is performed by using an accelerator opening sensor 1 in an injection amount characteristic map determined in advance by the accelerator opening and the engine speed.
1 is performed by obtaining a target fuel injection amount Qb corresponding to the current accelerator opening detected by the detected accelerator opening and the engine speed obtained from the BTDC signal and the like (the final fuel injection amount is determined by the necessary injection amount correction amount and Sometimes asked). (4) Further, based on the information obtained in the signal processing performed in S2, the timing at which each injector 3 should inject fuel,
That is, the fuel injection timing is calculated (S4). A target fuel injection timing corresponding to the fuel injection amount obtained in S3 and the current engine rotation speed is obtained from a fuel injection timing characteristic map determined in advance based on the fuel injection amount and the engine rotation speed. (5) Further, the fuel is injected based on the information obtained by the signal processing performed in S2 and so that the fuel injection amount determined in S3 can be injected at the fuel injection timing determined in S4. The pressure, that is, the fuel injection pressure is calculated (S
5). The control of the fuel injection pressure is performed by obtaining a target injection pressure from the fuel injection amount and the engine speed. More specifically, the common rail pressure Pcr is controlled by controlling a flow control valve provided in connection with the fuel pump 20. Is performed by controlling. The fuel injection control includes the injection amount control, the injection timing control, and the injection pressure control as described above, and the controller 37 controls the respective injectors 3 under the respective injection conditions.
The injectors 3 and the common rail pressure Pcr are controlled so that the fuel is injected from the injectors 3. After the CPU 14 is initialized in S1, S2 to S5 are sequentially executed for the injectors 3 that are to execute the fuel injection, and the above S2 to S5 are repeated for each injection. The fuel injection amount to be injected by the injector 3, the fuel injection timing, and the fuel injection pressure determined by the variable fuel pump 20 are optimized for fuel injection under the trade-off relationship between exhaust gas, noise, and output. Control.

One reference mark (missing tooth) is provided at an angular position corresponding to 120 ° before the compression top dead center of the reference cylinder # 1 on the rotary plate fixed to the pump shaft of the fuel pump 20 or the cam shaft for driving the intake and exhaust valves. May be formed), and the cylinder discriminating sensor 8
Outputs a REF signal once per rotation of the pump shaft by detecting this reference mark. When the cylinder discriminating sensor 8 outputs the REF signal, the REF signal as shown in FIG.
Interrupt processing is performed. FIG. 6 is a flowchart showing an interrupt process when a cylinder discrimination signal is input to the CPU 14 in the main process shown in FIG. In this interrupt processing, reset processing for setting the count value CNb of the BTDC signal to 0 is performed. That is, since the multi-cylinder engine 1 is a four-cylinder engine, the BTDC signal can take four integers from 0 to 3. Since the fuel injection and ignition in each cylinder makes a full cycle from the time when the count value CNb is 0, and before the count value CNb becomes 4, the cylinder discriminating sensor 8 detects this reference mark. It is set to 0 (S6).

The rotary plate fixed to the pump shaft of the fuel pump 20 or the cam shaft for driving the intake / exhaust valve has four marks before the top dead center (not shown) at angular positions corresponding to 60 ° before the compression top dead center in each cylinder. Teeth may be formed every 90 °.
When the DC sensor 9 detects the mark before the top dead center,
The BTDC signal is output four times per one rotation of the pump shaft (see the graph shown in the middle part of FIG. 3). BTDC signal is C
When input to the PU 14, the BTD
C signal interruption processing is performed.

The BTDC signal interrupt processing of FIG. 7 is performed as follows. FIG. 7 is a flowchart showing an interrupt process when a BTDC signal is input in the CPU main process shown in FIG. (1) The rotation speed of the engine 1 is calculated (S10).
That is, the current BTD after detecting the previous BTDC signal
The rotation speed of the engine 1 per unit time is calculated based on the time required until the detection of the C signal. (2) It is determined whether or not the count value CNb of the BTDC signal is 0 (S11). If the count value CNb is 0, the fuel injection process of the injector 31 provided in the cylinder (# 1) with the combustion order i = 1 (S2 to S5 and the subsequent fuel injection) is performed (S12). An outline of the timing of this injector processing is shown in the lowermost graph of FIG. (3) If the count value CNb is not 0 in the determination in S11, the process immediately proceeds to S13 to determine whether the count value CNb is 1 (S13). (4) If the count value CNb is 1, fuel injection processing (execution of S2 to S5 and subsequent fuel injection) of the injector 33 provided in the cylinder (# 3) of i = 2 is performed (S1).
4). (5) If the count value CNb is not 1 in the determination in S13, the process immediately proceeds to S15, and the same determination processing as described above and the injector fuel injection processing in the case of YES in the determination processing are performed (S15). ). (6) In S12, S14 or S15, # 1 to # 4
When the fuel injection processing of any of the injectors 3 is performed, the determination other than the determination of the corresponding count value CNb is always NO, so that the count value CNb which is increased by 1 in S16 is replaced with the new count value CNb. (S16)
This interrupt processing ends. Also in the next interrupt processing, the determination for the next count value CNb is made in S1.
YE in any of the similar determinations at S1, S13 or S15
It becomes S. When the count value CNb is sequentially increased and the count value CNb becomes 3, the reference mark is detected before the count value CNb becomes 4, and the count value CN is detected by S6.
b is reset to zero.

Next, the main processing of the DSP 15 will be described with reference to FIG. FIG. 8 is a flowchart showing DSP main processing in the controller shown in FIG. (1) Initialize the DSP (S20). (2) When initialization is completed, in-cylinder pressure processing is performed (S2).
1). The in-cylinder pressure processing is processing of in-cylinder pressure data of the detected cylinders # 1 to # 4. The in-cylinder pressure data processing is repeatedly performed for each of the corresponding cylinders # 1 to # 4 at every crank angle of 1 °. This is for calculating the generation rate and calculating the peak value of the pilot heat generation rate based on the calculation. Details of the in-cylinder pressure processing will be described later.

In the DSP, as a precondition for performing the in-cylinder processing shown in S21, an AD conversion end interrupt processing shown in FIG. 9 is performed. FIG. 9 is a flowchart showing an interrupt process at the end of AD conversion in the DSP main process shown in FIG. In this interrupt processing, the steps of reading the A / D conversion result of the in-cylinder pressure (that is, the combustion chamber pressure), initializing the crank angle, storing the in-cylinder pressure data in the memory, and updating the crank angle are performed by the crank angle Is executed in synchronization with the change of 1 °. That is, the in-cylinder pressure of each cylinder is converted into an AD converter 19 (FIG.
(See Figure 9)
The A / D conversion end interrupt processing shown in FIG.
Interrupt processing is performed each time the conversion is completed. (1) The AD conversion result ADr (i) of the in-cylinder pressure is read (S30). AD conversion result ADr (i) of each in-cylinder pressure
Is read as Pc (i) in the combustion order i (= 1 to 4). (2) Next, initialization of the crank angle is performed (S3).
1). (6) The in-cylinder pressure data is stored in the memory (S3
2). (7) The crank angle is updated (S33). S3
Details of each of 1 to S33 will be described below.

Next, the initialization of the crank angle in S31 will be described with reference to the flowchart shown in FIG. FIG. 10 is a flowchart showing the process of initializing the crank angle in the interrupt process at the end of the AD conversion shown in FIG. The engine is a four-cylinder engine, and the fuel injection and ignition of each cylinder makes one cycle with two rotations of the crankshaft.
0. The count value θ of the crank angle counts 1 at 1 °. The count value θ of the crank angle is 0 when the cylinder (# 1) having the combustion order i = 1 takes the position of the top dead center.
Until fuel injection and ignition of the four cylinders completes,
Take the count value up to 9. (1) It is determined whether or not the initialization of the crank angle has already been completed (S40). If the initialization of the crank angle has already been completed, the routine immediately returns to the AD conversion end interrupt routine. If the crank angle initialization has not been completed,
The step moves to S41. (2) It is determined whether or not the REF signal output at 120 ° before the top dead center of a specific cylinder, i.e., the cylinder with i = 1 (# 1), has risen (S41). If the REF signal has not risen, the process immediately returns to the AD conversion end interrupt routine. (3) If the REF signal has risen in S41, the count number θ of the crank angle is set to 720 minus 12
It is set to 0 (ie, 600) (S42). 120 is
This corresponds to the mounting angle of the cylinder discrimination sensor 8. (4) When the setting in S42 is completed, the initialization of the crank angle is completed (S43). Once the initialization of the crank angle is performed, the initialization of the crank angle is not performed unless the engine is started next time.

Next, the process of storing the in-cylinder pressure data in the memory in S32 will be described with reference to the flowchart shown in FIG. FIG. 11 is a flowchart showing a process of storing the in-cylinder pressure data in the memory in the interrupt process at the end of the AD conversion shown in FIG. (1) The count number θ of the initialized crank angle is 54
It is determined whether it is 0 or more or less than 180 (S5
0). If the count number .theta. Of the crank angle is not within this range, the flow shifts to S57. The range of the count number θ determined in S50 is a range in which the crank angle of the cylinder (# 1) of i = 1 is within 180 ° before and after the compression top dead center, and is a range from the compression stroke to the explosion stroke. The in-cylinder pressure during this time is important for ignition timing control and is adopted as data. (2) It is determined whether or not the count number θ of the crank angle is 540 (S51). This count number θ is 540
Is the starting point of the range of the count number θ, the cylinder-specific crank angle count number θc defined as the crank angle count number for the cylinder with i = 1
(1) is cleared to 0 (S52). 1 in parentheses
Indicates that it is the crank angle count number for the cylinder with i = 1. Therefore, θc (1) is from 0 to 3
Take values up to 59. (3) The in-cylinder pressure Pc (1) for the cylinder with i = 1 is
The in-cylinder pressure Pc (θc (1), 1) at the time of the cylinder-specific crank angle count number θc (1) is set (S53). (4) It is determined whether or not the cylinder-specific crank angle count number θc (1) for the cylinder with i = 1 is 359 (S54). That is, it is determined whether or not it is the end point of the data collection crank angle range for the cylinder with i = 1. (5) If θc (1) is 359, θc
Since (1) is after execution from 0 to 359, i
This means that the storage of the in-cylinder pressure data for the cylinder of = 1 has been completed (S55). (6) If θc (1) is not 359, θc (1)
Has not reached the end point yet, and in response to advancing the crank angle by 1 °, a value obtained by increasing the count of θc (1) by 1 is set as a new θc (1) (S5).
6).

(7) Next, the count number θ of the crank angle
Is greater than or equal to 0 and less than 360 (S5).
7). If the count .theta. Of the crank angle is not within this range, the flow shifts to S64. The range of the count number θ determined in S57 is a range in which the crank angle of the cylinder (# 3) with the combustion order i = 2 is within 180 ° before and after the compression top dead center, and the range from the compression stroke to the explosion stroke. It is. The in-cylinder pressure during this time is necessary for ignition timing control and is adopted as data. Therefore, when the count θ of the crank angle is not less than 0 and less than 180, i = 1
Of the cylinder (# 1) and the cylinder (# 3) with i = 2 are stored. (8) It is determined whether or not the count number θ of the crank angle is 0 (S58). If the count number θ is 0, it is the starting point of the range of the count number θ, and thus i =
Cylinder angle count number θc for each cylinder defined as the crank angle count number for the second cylinder (# 3)
(2) is cleared to 0 (S59). Therefore,
θc (2) for the cylinder with i = 2 also takes a value from 0 to 359. (9) The in-cylinder pressure Pc (2) for the cylinder of i = 2 at the time of the cylinder-specific crank angle count number θc (2) is set to the in-cylinder pressure Pc (θc (2), 2) (S60). (10) It is determined whether or not the cylinder-specific crank angle count number θc (2) for the cylinder with i = 2 is 359 (S61). (11) Assuming that θc (2) is 359,
Since c (2) is after executing from 0 to 359,
The storage of the in-cylinder pressure data for the cylinder with i = 2 ends (S62). (12) If θc (2) is not 359, θc
In (2), since the end point has not yet been reached, a value obtained by increasing the count number of θc (2) by 1 in response to advancing the crank angle by 1 ° is set as a new θc (2) (S63). ). (13) Thereafter, the same processing is performed for the cylinders of i = 3 and 4 (# 4 and # 2), and the in-cylinder pressure data is stored in the memory (S64).

Next, the process of updating the crank angle will be described with reference to the flowchart shown in FIG. FIG.
2 is a flowchart showing a crank angle update process in the interrupt process at the end of the AD conversion shown in FIG.
A crank angle update process S33 is performed for each increment of θc. (1) The count θ of the crank angle is increased by one and updated (S70). (2) It is determined whether or not the count number θ is less than 720 (S71). If the count number θ is less than 720, 4
Since the injection and ignition for one cylinder have not yet completed one cycle, the interrupt processing after the end of the AD conversion is continued. (3) If the count number θ is 720 or more, the count number θ is reset to 0 (S72).

Next, the details of the in-cylinder pressure process (S21) during the main process of the DSP shown in FIG. 8 will be described with reference to the flowchart shown in FIG. FIG.
9 is a flowchart showing in-cylinder pressure processing in the main processing of the DSP shown in FIG. 8. (1) Combustion order i = 1 performed in S55 of FIG.
It is determined whether the in-cylinder pressure data storage of the cylinder (# 1) has been completed (S80). If the storage of the in-cylinder pressure data of the cylinder (# 1) of i = 1 has not been completed, the flow shifts to S84 to determine whether the storage of the in-cylinder pressure data of the cylinder (# 3) of i = 2 has been completed. Is determined. (2) If it is determined in S80 that the storage of the in-cylinder pressure data of the cylinder with i = 1 has been completed, the data is subjected to a filtering process (S81). Since the in-cylinder pressure data greatly fluctuates, noise is removed by performing a filtering process such as taking a moving average to obtain a smooth in-cylinder pressure curve (see FIG. 2).

(3) From the obtained pressure curve, the heat release rate q
Is calculated (S82). The heat release rate q is obtained as follows. First, the in-cylinder volume Vθ is expressed by the following equation.

(Equation 1) Here, Vc is the clearance volume [m ³ ], S is the piston stroke [m], L is the connecting rod length [m], and θc is the crank angle (deg). The in-cylinder volume Vθ and its differential value based on the crank angle θc may be calculated in real time each time the crank angle θc is updated, but may be sequentially calculated from map data obtained in advance and stored in a memory. You may. The cylinder pressure Pc and its differential value based on the crank angle θc are obtained by detecting with a sensor and processing with a DSP. Heat release rate q = dQ / dθ
Is calculated by the following equation.

(Equation 2) Here, assuming that the specific heat ratio κ is constant, the above equation is calculated in real time. (4) Based on the result of the heat release rate q calculated in S82, the heat release rate generated by the pilot injection, that is, qmax (1), which is the peak value of the pilot heat release rate, is calculated (S83). Peak value of pilot heat release rate q
Details of the calculation of max (1) will be described later.

(5) If the in-cylinder pressure data storage of the cylinder (# 1) with the combustion order i = 1 has not been completed, is the in-cylinder pressure data storage of the cylinder (# 3) with i = 2 completed? It is determined whether or not it is (S84). Filter processing after S84 (S8
5), the calculation of the heat release rate q (S86), and the calculation of the peak value qmax (2) of the pilot heat release rate (S87) are the same as steps S80 to S83, and will not be described.
Similarly, cylinders with i = 3 (# 4) and cylinders with i = 4 (#
Similar processing is performed for 2). The processing in FIG. 13 is performed in a state where the in-cylinder pressure data for 10 combustion strokes of each cylinder has been stored by the interruption of the crank angle, and the filtering processing, the calculation of the heat release rate, the pilot heat release rate peak value q
Each step of the calculation of max (i) is performed.

As described above, the fuel injection amount characteristic map determined in advance based on the accelerator opening, the engine speed, and the like is used to determine the target corresponding to the accelerator opening, the engine speed, etc., which is the current engine operating state. Fuel injection quantity Q
In the case of performing pilot injection, a pilot fuel injection amount (hereinafter abbreviated as pilot injection amount) Qp is further obtained from the target fuel injection amount Qb and the engine speed to obtain the target fuel injection amount. Qb is divided into a pilot injection amount Qp and a main fuel injection amount (hereinafter abbreviated as main injection amount) Qm. There is a correlation as shown in FIG. 18 between the pilot injection amount Qp and the pilot heat release rate peak value qmax. That is, when the ratio of the pilot injection amount Qp to the target fuel injection amount Qb is increased, the pilot heat release rate peak value qmax increases and the heat release rate peak due to the main injection decreases. Among the heat release rates q shown in FIG. 18, it is most desirable to perform the pilot injection so as to cause the change shown by the broken line. When the engine is in an idling operation state, the pilot injection pulse width for driving the electromagnetic actuator 16 of the injector 3 is corrected so that the pilot heat release rate peak value becomes a preset target peak value. Even if the characteristics fluctuate or change over time, the optimum pilot injection and its combustion are performed.

Next, the calculation of the pilot heat release rate peak value qmax (i) will be described with reference to FIG.
FIG. 14 is a flowchart showing a process of calculating the pilot heat release rate peak value qmax (i) in the in-cylinder pressure process shown in FIG. i is a combustion number that can take a value of 1 to 4, and a common process is performed for each value of i. (1) The start timing (hereinafter abbreviated as pilot injection timing) Tp of a pulse supplied to the electromagnetic actuator 16 of the injector 3 for performing the pilot injection, and the electromagnetic actuator 1 of the injector 3 for performing the main injection.
The start timing (abbreviated as main injection timing) Tm of the pulse supplied to the motor 6 and the engine rotation speed Ne are read from the dual port memory 16 (S90). These data are processed by the CPU 14 and written in the dual port memory 16 which is a RAM. (2) The start delay θa (crank angle) of the calculation period representing the delay from the pilot injection timing Tp to the timing of starting the calculation of the pilot heat release rate peak value qmax (i) is calculated (S91). As shown in FIG. 19, the start delay θa of the calculation period is determined in advance to be a value obtained by adding the ignition delay α to the injection delay θi of the injector 3 or to be smaller than the value. The injection delay θi is substantially constant over time, and the ignition delay α changes depending on the engine rotation speed Ne. Therefore, the start delay θa of the calculation period is map data using the engine rotation speed Ne as a variable. (3) From the pilot injection timing Tp and the start delay θa,
The calculation start angle θs of the pilot heat release rate peak value qmax is calculated by the following equation (S92). Note that the dimension of the pilot injection timing Tp is adjusted to the crank angle. θs ← Tp + θa (4) From the main injection timing Tm, the calculation period end delay θb of the pilot heat release rate peak value qmax is calculated (S
93). The calculation period end delay θb is defined as the injection delay from the main injection timing Tm corresponding to the rise of the drive pulse supplied to the electromagnetic actuator of the injector for the main injection to the injection of the main injection, and the injection of the main injection. It is calculated as a period corresponding to the sum of the ignition delay β and the time at which the fuel ignites. Since the ignition delay β depends on the engine rotation speed Ne, the calculation period end delay θb
Is map data using the engine speed Ne as a variable. (5) From the main injection timing Tm and the calculation period end delay θb, the pilot heat release rate peak value qma is calculated by the following equation.
The calculation end angle θe of x is calculated (S94). Note that the dimension of the pilot injection timing Tp is adjusted to the crank angle. θe ← Tm + θb (6) The pilot heat release rate peak value qmax as the maximum value during the calculation angle period from θs to θe is obtained by calculation and written in the dual port memory 16 (S9).
5). The relationship between the calculation section start delay θa, the calculation section end delay θb, and the engine rotation speed is shown in the graph of FIG. Generally, as the engine speed increases, the crank angle that appears as a delay tends to decrease.

The calculation of the pilot heat release rate peak value qmax shown in S95 will be described with reference to FIG. FIG. 15 is a flowchart showing the calculation of the pilot heat release rate peak value qmax (i) in the calculation of the pilot heat release rate peak value qmax shown in FIG. (1) The start point θs of the calculation angle period is set as θmax, and the heat release rate q (θs) at the start point θs is set as qmax (S
100). (2) A new crank angle θ is obtained by adding an increment of 1 ° to the crank angle θ (S101). (3) Heat release rate q (θ) at new crank angle θ
Is larger than the previous heat release rate q (θ-1) (S102). (4) When the heat release rate q (θ) is larger than the heat release rate q (θ-1), qmax is replaced by the heat release rate q (θ) at a new crank angle θ,
θmax is replaced with the new crank angle θ (S10
3). (5) When the heat release rate q (θ) is equal to or less than the heat release rate q (θ−1) in the determination in S102, or when the replacement in S103 is completed, the crank angle θ is set to be equal to or less than the end point θe of the calculation angle period. It is determined whether there is (S104). If the crank angle θ has not yet reached the end point θe of the calculation angle period, the process returns to S101 and the above processing is repeated.
When the crank angle θ reaches the end point θe of the calculation angle period, the flow returns to the calculation flow of the pilot heat release rate peak qmax shown in FIG.

Next, the pilot heat release rate peak value qmax will be described with reference to the flowcharts shown in FIGS.
When (i) is obtained, the pilot heat release rate peak value qma
The drive pulse width supplied to the electromagnetic actuator 16 of each injector 3 is corrected so that x (i) matches a predetermined target value. The correction process is called the combustion order i
This will be described below by taking as an example the injector 31 provided in the cylinder # 1 where is 1. Note that the combustion order i is limited to 1 here, and is omitted for simplification of display. FIG. 16 is a flowchart showing the correction processing of the pilot injection driving pulse width in the driving processing of the injector 31 in the BTDC signal interruption processing shown in FIG. As a premise of the correction processing of the pilot injection drive pulse width, the REF interrupt processing is started in synchronization with the cylinder determination REF signal shown in FIG.
It is assumed that Nb is reset, the BTDC signal interruption process is started in synchronization with the BTDC signal shown in FIG. 7, and the cylinder to inject fuel is determined according to the value of the cylinder determination counter CNb. (1) The fuel injection condition for injecting fuel by the corresponding injector is read (S110). The fuel injection condition is the fuel injection pressure (that is, the common rail pressure P
cr), main fuel injection amount Qm, pilot fuel injection amount Qp, main injection timing Tm, and pilot injection timing T
p. (2) It is determined whether the pilot injection amount Qp is currently 0 (S111). When the pilot injection amount Qp is 0, there is no need to find the pilot heat release rate peak value qmax, so the process proceeds to S119 and the drive supplied to the electromagnetic actuator 66 (see FIG. 21) of the injector 31 to perform the main injection. The pulse width (hereinafter abbreviated as main injection pulse width) is obtained. (3) If the pilot injection amount Qp is not 0, it is determined whether or not the operation state of the engine is an idle operation state (S112). If the operation state of the engine is not the idling operation state, the problems (instability of operation, noise, exhaust gas characteristics) associated with the idling operation of the engine state do not become particularly noticeable. Subsequently, the fuel injection control involving the pilot injection is performed. (4) When the operation state of the engine is the idling operation state, for example, the target pilot heat release rate peak value qbmax according to the pilot injection amount Qp and the pilot injection timing Tp at that time is calculated (S113).

(5) The injection cylinder (#) obtained in S103 and stored in the dual port memory 16 (see FIG. 4)
Deviation Δqmax between the pilot heat release rate peak qmax of 1) and the target pilot heat release peak qbmax.
Is calculated (S114). That is, Δqmax = qbma
x-qmax (6) The electromagnetic actuator 66 of the injector 31 (see FIG.
Suppose that a correction amount ΔPWp of the drive pulse width (hereinafter, abbreviated as pilot injection pulse width) PWp supplied to the drive pulse width PWp is supplied. As in the following equation, the value obtained by multiplying the gain by the deviation Δqmax obtained in S114 is added to the previously obtained correction amount ΔPWp of the pilot injection pulse width PWp to obtain the current pilot injection pulse width correction amount ΔPWp. (S115). By this processing, integral control of the pilot injection pulse width can be performed. ΔPWp ← ΔPWp + Δqmax × Gp where Gp is a gain. (7) The pilot injection pulse width PWp is determined from the fuel injection pressure Pcr, the pilot injection amount Qp, and the injector flow rate characteristics (S116). Subsequent processing is also performed when the engine is not in idle operation. (8) Pilot injection pulse width PWp obtained in S116
The final pilot injection pulse width PWpf is obtained by adding the current pilot injection pulse width correction amount ΔPWp obtained in S115 (S117). PWpf ← PWp + ΔPWp (9) A pilot injection drive pulse output counter (not shown) that outputs a drive pulse for driving the electromagnetic actuator of the injector 31 for pilot injection.
The pilot injection timing Tp read in S110 and the final pilot injection pulse width PWp obtained in S117
The injection pulse width corresponding to f is set (S118). (10) Since the pilot injection pulse width PWp is corrected, an injection amount different from the pilot injection amount Qp read in S110 is actually injected. Therefore, in order not to change the total fuel injection amount as the sum of the pilot injection and the main injection, the fuel injection pressure and the final pilot injection amount corresponding to the final pilot injection pulse width PWpf are used.
The main injection pulse width PWm is calculated (S119). (11) A main injection drive pulse output counter (not shown) that outputs a drive pulse for driving the electromagnetic actuator 66 of the injector 31 for main injection.
The main injection timing Tm read in S110 and S
An injection pulse width corresponding to the main injection pulse width PWm obtained in 119 is set (S120).

As described above, the drive pulse width for pilot injection is corrected so that the calculated pilot heat release rate peak value qmax (i) matches the target pilot heat release rate peak qbmax (i). In the pilot injection, an optimum pilot heat release rate peak value qmax according to the operating state of the engine at that time can be obtained. Therefore, it is possible to prevent phenomena such as unstable operation of the engine, deterioration of engine noise and exhaust gas performance, particularly when the engine is in an idle operation state.

Next, the calculation of the actual fuel ignition timing Td (i) (FIG. 2) using the data of the heat release rate will be described with reference to FIG. FIG. 17 shows the actual fuel ignition timing Td based on the heat release rate obtained in the in-cylinder pressure process shown in FIG.
It is a flowchart which shows the calculation process which calculates | requires (i).
A common process is performed for each combustion order i (possible values of 1 to 4). Θc is a count value of a crank angle in the compression and expansion process of each cylinder, and counts a crank angle in a range of 180 ° before and after the top dead center of the explosion stroke from 0 to 359. This flowchart is executed when θc is 3 or more, as shown in S130. S83 and S8
6 and the like, the heat release rate q has already been determined. Also,
Reference is made to the graph of the heat release rate q with respect to the crank angle count value θc shown in FIG. The actual fuel ignition timing T
d (i) is data on the dual port memory 16 (see FIG. 4). (1) It is determined whether or not q (i) (θc−3) is negative for the combustion order i at that time, that is, for the cylinder in which fuel ignition and combustion are being performed (S131). (2) If q (i) (θc−3) is negative, whether q (i) (θc−2) when the count value θc of the crank angle of the cylinder advances by one count is negative. Is determined (S132). (3) If q (i) (θc−2) is negative, is q (i) (θc−1) positive when the count value θc of the crank angle of the cylinder is further advanced by one count? It is determined whether or not it is (S133). (4) If q (i) (θc-1) is positive, it is determined whether q (i) (θc) when the count value θc of the crank angle of the cylinder is further advanced by one count is positive. Is determined (S134).

(5) If q (i) (θc) is positive, it is assumed that the actual fuel ignition timing Td (i) is θc−2 (S
135). In other words, since the sign of the heat release rate q changes between θc−2 and θc−1 at the intermediate point in the count value of four consecutive crank angles, the actual fuel ignition timing Td (i) Is regarded as θc-2. Actually, the zero cross point is between θc−2 and θc−1, so the zero cross point may be obtained by interpolation. If the calculation accuracy is good, a point at which the sign changes from negative to positive at two points before and after may be determined as the zero cross point. (6) When q (i) (θc-3) is positive in S121, S
When q (i) (θc-2) is positive in 132, q (i) (θc-1) is negative in S133, and q in S134
(I) When (θc) is negative, the actual fuel ignition timing Td (i) is set to the crank angle count ignition 0 timing (S136). That is, only when the sign of q changes between θc−2 and θc−1, the actual fuel ignition timing T
d (i) is obtained, and in all other situations, the actual fuel ignition timing Td (i) is set to 0. (7) It is determined whether or not the actual fuel ignition timing Td (i) is 0 (S137). (8) If the actual fuel ignition timing Td (i) is 0, the count value θc of the crank angle is increased by 1 (S13).
8). (9) It is determined whether or not the count value θc of the crank angle is 360 (S139). That is, it is determined whether or not the count value is the last count value in the range that θc can take. If θc is not yet 360, the process returns to S131 and executes the routine again.

Referring again to FIG. 2, when fuel injection is performed at time T0 and the injected fuel is ignited, the rise in the in-cylinder pressure Pc becomes slightly gentle, and then increases greatly by the subsequent ignition. Although it has been difficult to accurately determine the crank angle at which fuel is ignited from the rapidly rising in-cylinder pressure Pc, it is difficult to accurately determine the actual fuel ignition timing as described above by focusing on the heat release rate q. it can. That is, FIG.
When the fuel is injected into the combustion chamber at the fuel injection rate Rf shown in the middle graph, as shown in the lower graph of heat release rate q, a negative heat release rate is once exhibited by heat absorption, but thereafter,
Change to positive heat release rate. At the beginning of the fuel injection, the heat release rate q becomes slightly negative (due to fuel vaporization), but starts to increase due to the fuel ignition, and the zero-crossing time T1 at which the heat release rate q changes from negative to positive is defined as fuel ignition. The actual fuel ignition timing can be easily determined by considering the timing Td (the ignition timing of the fuel injected by the pilot injection when performing the pilot injection). In addition, from time T0 to time T
The period up to d corresponds to the fuel ignition delay period. When the actual fuel ignition timing Td (i) is determined, the fuel injection timing in each injector is corrected so that the actual fuel ignition timing matches the target fuel ignition timing determined according to the operating state. The specific correction processing of the fuel injection timing is not the subject of the present invention, and therefore will be omitted.

The contents of the above flowchart are summarized with reference to FIG. 3. The four graphs in the upper part of FIG. 3 show changes in the in-cylinder pressure Pc with the passage of the crank angle θ of the cylinders # 1 to # 4. Represents. # 1 cylinder, # 3 cylinder, # 4
The compression explosion stroke will follow one after another in the order of cylinder and cylinder # 2.
When one cylinder goes through an explosion stroke, the next cylinder is in a compression stroke. In addition, 60 ° before compression top dead center of each cylinder
Outputs a BTDC signal. The count value θ of the crank angle is 0 at the compression top dead center of the # 1 cylinder. Since the engine is a four-cycle engine, the crankshaft makes two revolutions, that is, one cycle with the count value 719. For the # 1 cylinder, 180 ° before the compression top dead center, that is, when the count value is 540 or more,
180 ° after compression top dead center, ie, 180
If it is less than the cylinder pressure, the in-cylinder pressure data of cylinder # 1 is stored. Based on the stored data, a menn process is performed within a predetermined time from 180 ° after the compression top dead center, and the next BT
The injection valve driving process of the injector 31 is performed based on the DC interrupt signal. Further, a heat release rate peak qmax due to ignition and combustion of the fuel injected by the pilot injection in the calculation section as shown in FIG. 19 is obtained, and as shown in FIG. 18, a target optimum pilot heat release rate peak value is obtained. Is maintained, the pilot injection timing and the drive pulse width of the electromagnetic actuator of the injector are controlled.

[0056]

The method and the device for controlling the pilot injection quantity of the engine according to the present invention have the following effects because they are configured as described above. That is, even if there is an individual difference between the individual injectors, or even if the same injector has a case where the pilot injection amount characteristic fluctuates due to aging, the pilot heat release rate peaks at the time of idle operation or the like. The drive pulse width of the electromagnetic actuator of the injector is corrected so as to have a target value corresponding to the operating state. Since the variation of the pilot injection amount is suppressed and the fuel injected by the pilot injection and the ignition are performed while maintaining the optimum pilot injection amount, the amount of NOx or smoke contained in the exhaust gas may be increased. Absent. Particularly, in the idling operation state of the engine, the operating state of the engine originally tends to be unstable because the engine speed is low. However, in the method and apparatus for controlling the pilot injection amount of the engine according to the present invention, the pilot injection fuel is controlled. Since the variation in the amount is reduced, when the combustion by the pilot injection is stabilized, the combustion by the subsequent main fuel injection is also stabilized, and a good operating state of the engine is ensured, so that the exhaust gas performance is maintained satisfactorily. As a result, engine operation does not become unstable, engine noise is suppressed, and N
Deterioration of exhaust gas performance such as generation of Ox can be prevented.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing an example of a common rail fuel injection system including a multi-cylinder engine to which a pilot injection amount control method and device according to the present invention are applied.

FIG. 2 is a graph showing how the in-cylinder pressure, fuel injection rate, and heat generation rate change with the passage of the crank angle of the engine shown in FIG.

FIG. 3 is a graph showing an outline of in-cylinder pressure, data processing, and the like in an injector according to progress of a crank angle of the multi-cylinder engine shown in FIG.

FIG. 4 is a block diagram showing an outline of a controller for controlling a pilot injection amount of an engine according to the present invention.

FIG. 5 is a flowchart showing a CPU main process in the controller shown in FIG. 4;

FIG. 6 shows a REF in the CPU main processing shown in FIG. 5;
It is a flowchart which shows the interruption processing when a signal is input.

FIG. 7 shows a BTD in the CPU main processing shown in FIG. 5;
It is a flowchart which shows the interruption processing when a C signal is input.

FIG. 8 is a flowchart showing a DSP main process in the controller shown in FIG. 4;

FIG. 9 is a flowchart showing an interrupt process at the end of AD conversion in the DSP main process shown in FIG. 8;

FIG. 10 is a flowchart showing a crank angle initialization process in the interrupt process at the end of the AD conversion shown in FIG. 9;

11 is a flowchart showing a process of storing in-cylinder pressure data in a memory in an interrupt process at the end of AD conversion shown in FIG. 9;

12 is a flowchart showing a crank angle update process in the interrupt process at the end of the AD conversion shown in FIG. 9;

FIG. 13 is a flowchart showing in-cylinder pressure processing by a DSP in the controller shown in FIG. 4;

14 is a flowchart showing a process of calculating a pilot heat release rate peak value in the in-cylinder pressure process shown in FIG.

FIG. 15 is a flowchart showing a process of calculating a pilot heat release rate peak value during a calculation period in the process of calculating the pilot heat release rate peak value shown in FIG. 14;

FIG. 16 is a flowchart showing an injector process in the BTDC signal interrupt process shown in FIG. 7;

FIG. 17 is a flowchart showing a calculation process for obtaining an actual fuel ignition timing from a heat release rate.

FIG. 18 is a graph showing a change in a heat generation rate with the passage of a crank angle.

FIG. 19 is a graph showing changes in the injector drive pulse, the needle valve lift of the injector, and the heat generation rate between the pilot injection and the main injection.

FIG. 20 is a graph showing a relationship between an engine rotation speed, a calculation period start delay θa, and a calculation period end delay θb.

FIG. 21 is a longitudinal sectional view showing a conventional injector.

FIG. 22 is a graph showing a relationship between an injection pulse width and a fuel injection amount using a fuel pressure as a parameter in the injector shown in FIG. 21;

FIG. 23 is an explanatory diagram showing a relationship between a pilot injection drive pulse width and a variation in pilot injection amount.

FIG. 24 is a graph illustrating detection of the actual fuel ignition timing based on the flowchart shown in FIG. 17;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Multi-cylinder engine 2 Cylinder 3, 31-34 Injector 4, 41-44 In-cylinder pressure sensor 8 Cylinder discrimination sensor 9 BTDC sensor 10 Crank angle sensor 11 Accelerator opening sensor 12 Common rail pressure sensor 22 Common rail 23 Branch pipe 37 Controller 61, 62 Fuel passage 64 Needle valve 65 Injection hole Pc In-cylinder pressure θ Crank angle count value θc Crank angle count value per cylinder q Heat release rate qmax Peak pilot heat release rate qbmax Target peak heat release rate i Fuel Order Qb Target fuel injection amount Qp Pilot injection amount Qm Main injection amount Tp Pilot injection timing Tm Main injection timing Td Actual fuel ignition timing PWp Pilot drive pulse width ΔPWp Correction of PWp PWpf Final pilot drive pulse width

Claims (10)

[Claims] 1. A fuel injected from an injector into a combustion chamber is divided and injected into a main injection and a pilot injection preceding the main injection, and the fuel in the combustion chamber is detected based on the sequentially detected pressure in the combustion chamber. A heat release rate is calculated, and a peak value of a pilot heat release rate caused by the ignition of the fuel injected in the pilot injection is calculated from the heat release rate, and a peak value is determined in advance based on an operation state of the engine. Determining the target peak value of the pilot heat release rate according to the current operating state of the engine from the target peak value of the pilot heat release rate, and calculating the peak value and the target peak value of the pilot heat release rate. Correction of the pilot fuel injection amount injected by the pilot injection so as to eliminate the deviation of the engine. Control method for pilot fuel injection quantity. 2. The volume in the combustion chamber is determined sequentially, and the heat release rate in the combustion chamber is determined based on the detected pressure in the combustion chamber, the determined volume in the combustion chamber, and changes in the volume. The method according to claim 1, wherein the pilot fuel injection amount is calculated from the rate. 3. The peak value of the pilot heat release rate is calculated from a pilot start timing at which a pilot injection drive pulse is supplied to the injector to perform the pilot injection, and a calculation start timing that elapses from the pilot injection timing. It is determined as the maximum value of the heat release rate in a calculation period between a main injection timing at which a main injection drive pulse is supplied to the injector to perform the main injection and a calculation end time at which an end delay has elapsed. A method for controlling a pilot fuel injection amount in an engine according to claim 1 or 2. 4. The start delay is defined as an injection delay of the injector from the pilot injection timing until the injector actually starts the pilot injection and a time until the fuel injected as the pilot injection from the injector ignites. The end delay is the injection delay of the injector from the main injection timing until the injector actually starts the main injection, and the injection is performed as the main injection from the injector. 4. The control method according to claim 3, wherein the delay is a sum of an ignition delay until the fuel is ignited. 5. The fuel supplied to the injector is lifted by a needle valve in response to the driving of an electromagnetic actuator provided in the injector, whereby an injection hole formed at the tip of the injector is opened. The pilot injection quantity in an engine according to any one of claims 1 to 4, wherein the pilot fuel injection quantity is performed, and the pilot fuel injection quantity is corrected by changing a drive pulse width of the injector to the electromagnetic actuator. Control method. 6. An injector for injecting fuel supplied through a fuel flow path into a combustion chamber of an engine, and an injector for dividing and injecting the fuel into a main injection and a pilot injection preceding the main injection. A controller for controlling driving, wherein the controller calculates a heat release rate in the combustion chamber based on the sequentially detected pressure in the combustion chamber, and calculates a rate of the fuel injected in the pilot injection from the heat release rate. Calculate the peak value of pilot heat release rate caused by ignition,
Calculating a target peak value of the pilot heat generation rate corresponding to the current operation state of the engine from a target peak value of the pilot heat generation rate determined in advance based on an operation state of the engine; A pilot fuel injection amount control device for an engine, comprising: correcting a pilot fuel injection amount injected by the pilot injection so as to eliminate a deviation between the peak value of the occurrence rate and the target peak value. 7. The controller according to claim 1, wherein the controller sequentially calculates a volume in the combustion chamber, and calculates the volume in the combustion chamber based on the sequentially detected pressure in the combustion chamber, the volume in the combustion chamber, and a change rate thereof. 7. The pilot fuel injection amount control system for an engine according to claim 6, comprising calculating a heat release rate. 8. The controller according to claim 1, wherein the controller calculates the peak value of the pilot heat generation rate after a start delay from a pilot injection timing at which a pilot injection drive pulse is supplied to the injector to perform the pilot injection. It is determined as the maximum value of the heat release rate in a calculation period between a timing and a calculation end time when an end delay has elapsed from a main injection time at which a main injection drive pulse is supplied to the injector to perform the main injection. The pilot fuel injection amount control device for an engine according to claim 6 or 7, further comprising: 9. The delay of the start is defined as an injection delay of the injector from the pilot injection timing until the injector actually starts the pilot injection and a time until the fuel injected as the pilot injection from the injector ignites. The end delay is the injection delay of the injector from the main injection timing until the injector actually starts the main injection, and the injection is performed as the main injection from the injector. 9. The pilot fuel injection amount control device for an engine according to claim 8, wherein the delay is a sum of an ignition delay until the fuel is ignited. 10. The injector, comprising: an electromagnetic actuator; a needle valve which is lifted in response to driving of the electromagnetic actuator; and an injection hole formed at a tip of the injector and opened by lifting the needle valve. 7. The method according to claim 6, wherein the pilot fuel injection amount is corrected by changing a drive pulse width of the injector to the electromagnetic actuator.
A pilot injection amount control device for an engine according to any one of claims 9 to 9.