JP2018076804A - Control device of internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control device of an internal combustion engine which can predict a fuel pressure value on the basis of a detection value of a fuel pressure sensor in which an actual fuel pressure value is reflected.SOLUTION: A control device of an internal combustion engine comprises: a high-pressure pump for generating the pulsation of fuel pressure in a low-pressure fuel passage; a center fuel pressure calculation section for calculating a center fuel pressure value Pof the pulsation on the basis of a this-time value being a detection value of the fuel pressure sensor which is acquired this time; an amplitude calculation section for calculating an amplitude A of the pulsation on the basis of a last-time value being a detection value which is acquired last time, and the this-time value; a phase calculation section for calculating an initial phase B of the pulsation on the basis of one of the last-time value and the this-time value; a storage section in which the number c of times of the discharge of fuel by the high-pressure pump per crank angle 360° is stored; and a prediction section for predicting a fuel pressure value P (θ) corresponding to a crank angle θ on the basis of the calculated center fuel pressure value P, the amplitude A, the initial phase B, the number c of times of the discharge, and a model formula.SELECTED DRAWING: Figure 11

Description

  The present invention relates to a control device for an internal combustion engine.

  An internal combustion engine having an in-cylinder injection valve and a port injection valve is known. In such an internal combustion engine, the fuel sucked up by the low pressure pump is supplied to the port injection valve through the low pressure fuel passage, and the fuel further pressurized by the high pressure pump is supplied to the in-cylinder injection valve through the high pressure fuel passage. Is done. In such a configuration, a fuel pressure sensor that detects the fuel pressure in the low-pressure fuel passage may be provided. The detection value of such a fuel pressure sensor may be used for control of the fuel injection amount from various control objects, for example, a port injection valve.

  In the above configuration, the fuel pressure may pulsate in the low pressure fuel passage mainly due to driving of the high pressure pump. If the fuel injection amount of the port injection valve is controlled based on the detection value of the fuel pressure sensor while pulsation is occurring, the fuel injection amount may not be appropriately controlled. The reason is as follows. A predetermined period is required after the detection value of the fuel pressure sensor is acquired until the port injection is started based on the detection value. Therefore, if pulsation occurs, the fuel pressure fluctuates greatly between the detection value being acquired and the start of port injection, and the acquired detection value and the actual fuel pressure value during port injection This is because there may be cases where the values differ greatly.

  On the other hand, for example, in Patent Document 1, the actual fuel pressure value is not used during the occurrence of pulsation, but the correction value of the port injection amount is determined based on a map in which the correction value of the port injection amount is defined in advance for each rotation speed of the internal combustion engine. A technique for controlling the fuel injection amount of the injection valve is described.

JP 2012-237274 A

  However, it is considered that the actual fuel pressure value is also influenced by the load and temperature of the internal combustion engine, the characteristics of the fuel used, and the like, which are factors other than the rotational speed of the internal combustion engine. For this reason, even if the fuel injection amount that is the control target is controlled based on the map as described above, the fuel injection amount may not be appropriately controlled. In addition, it is difficult to create a map associated with each of a plurality of other factors including the rotational speed of the internal combustion engine.

  Therefore, even if pulsation is occurring, it is desirable to control the control target based on the detection value of the fuel pressure sensor that reflects the actual fuel pressure value. However, as described above, the detection value of the fuel pressure sensor and the control target are controlled. Since the fuel pressure value during the process may differ greatly, there is a possibility that the controlled object cannot be controlled appropriately.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a control device for an internal combustion engine that can accurately predict a fuel pressure value based on a detection value of a fuel pressure sensor that reflects an actual fuel pressure value.

The above object is achieved by an in-cylinder injection valve that directly injects fuel into the cylinder of the internal combustion engine, a port injection valve that injects fuel into the intake port of the internal combustion engine, a low-pressure pump that pressurizes the fuel, and the low-pressure pump. A low-pressure fuel passage that supplies pressurized fuel to the plurality of port injection valves, and a fuel that is driven in conjunction with the internal combustion engine, further pressurizes the fuel supplied from the low-pressure fuel passage, and has a fuel pressure in the low-pressure fuel passage. A high-pressure pump that generates pulsation of the fuel, a high-pressure fuel passage that branches from the low-pressure fuel passage and that supplies fuel pressurized by the high-pressure pump to the plurality of in-cylinder injection valves, and a crank angle of the internal combustion engine is detected A crank angle sensor that detects the fuel pressure in the low-pressure fuel passage, an acquisition unit that acquires detection values of the fuel pressure sensor at regular sampling time intervals, and at least the acquisition Based on the previous value and present value is the detected value obtained last time and this time the parts, the center fuel pressure calculating section for calculating a center fuel pressure value P _c of the pulse _[kPa], at least, the previous value and the current An amplitude calculation unit that calculates the amplitude A [kPa] of the pulsation based on the value, and a phase calculation unit that calculates the initial phase B [deg] of the pulsation based on at least one of the previous value and the current value A storage unit that stores the number c of fuel discharges by the high-pressure pump per 360 degrees of crank angle, the calculated central fuel pressure value P _c , the amplitude A, the initial phase B, and the number of discharges c and a prediction unit that predicts a fuel pressure value P (θ) [kPa] corresponding to the crank angle θ [deg] based on the following equation (1), and the phase calculation unit Value and this time The one of the values, the crank angle at the time when the one of the previous value and the current value is acquired, the number of discharges c, the calculated amplitude A, and the calculated central fuel pressure value P _c. And calculating the initial phase B based on
The phase calculating unit calculates the one of the previous value and the current value, the crank angle at the time when the one of the previous value and the current value is acquired, and the number of discharges c. Based on the amplitude A, a candidate calculation unit that calculates two candidates of the initial phase B, the calculated amplitude A and the central fuel pressure value _Pc , one of the two candidates, the previous value, and the A temporary fuel pressure value calculating unit that calculates a temporary fuel pressure value by substituting the crank angle at the time when the one or the other of the current values is acquired and the number of discharges c into the equation (1); When the absolute value of the difference between the temporary fuel pressure value and the previous value or the current value corresponding to the temporary fuel pressure value is less than a predetermined value, the one of the candidates is specified as the initial phase B, and If the absolute value of the difference exceeds the predetermined value, the candidate This can be achieved by a control device for an internal combustion engine including a candidate specifying unit that specifies the other as the initial phase B.

  ADVANTAGE OF THE INVENTION According to this invention, the control apparatus of the internal combustion engine which can estimate a fuel pressure value based on the detected value of the fuel pressure sensor in which the actual fuel pressure value was reflected can be provided.

FIG. 1 is a schematic configuration diagram of an engine system according to the present embodiment. FIG. 2 is a waveform diagram of fuel pressure. FIG. 3 is a graph showing an example of the waveform of the fuel pressure pulsation and the injection period by the port injection valve. FIG. 4A is a graph showing a plurality of detected values acquired by the ECU, and FIG. 4B is a graph showing modeled fuel pressure pulsation. FIG. 5 is a flowchart illustrating an example of port injection control. FIG. 6 is a flowchart showing a first example of the central fuel pressure calculation process. FIG. 7 is a flowchart showing a second example of the central fuel pressure calculation process. FIG. 8 is a graph showing acquired detection values and spoofed values. FIG. 9 is a flowchart showing a third example of the central fuel pressure calculation process. FIG. 10 (A) is a graph showing the relationship between the number of samples of the detected value used for calculating the central fuel pressure value and the engine speed. FIGS. 10 (B) and 10 (C) show the fuel pressure pulsation. It is a graph which showed the number of samples. FIG. 11 is a flowchart illustrating an example of the phase calculation process. FIG. 12 is a graph showing a case where the model fuel pressure is delayed with respect to the actual fuel pressure.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic configuration diagram of an engine system 1 (hereinafter referred to as a system) of the present embodiment. The system 1 includes an engine 10 and an ECU (Electronic Control Unit) 41 that controls the engine 10. The engine 10 is a spark ignition type in-line four-cylinder engine including a cylinder group 11 including cylinders 111 to 114 arranged in series, an in-cylinder injection valve group 37, and a port injection valve group 27. The engine 10 is an example of an internal combustion engine, and the system 1 is an example of a control device for the internal combustion engine.

  The in-cylinder injection valve group 37 includes in-cylinder injection valves 371 to 374 that inject fuel into the cylinders 111 to 114, respectively. The port injection valve group 27 includes port injection valves 271 to 274 for injecting fuel into the intake ports 13 communicating with the cylinders 111 to 114, respectively. Each of the in-cylinder injection valve group 37 and the port injection valve group 27 is an electromagnetically driven on-off valve in which the fuel injection amount is adjusted by energizing the electromagnetic coil in a predetermined energization period to separate the valve body from the valve seat. It is.

  The engine 10 is formed with an intake passage 12 having a plurality of intake ports 13 corresponding to each of the cylinder groups 11 and an exhaust passage having a plurality of exhaust ports (not shown). In each of the cylinder groups 11, a piston (not shown) is accommodated to define a combustion chamber. The combustion chamber is opened and closed by an intake valve and an exhaust valve. Further, the engine 10 includes a spark plug (not shown). The engine 10 includes a crankshaft 14 that is linked to a plurality of pistons, and a camshaft 15 that is linked to the crankshaft 14 and drives an intake valve or an exhaust valve. Further, a crank angle sensor 14a for detecting the rotation angle of the crankshaft 14 is provided. The resolution of crank angle detection by the crank angle sensor 14a is preferably a high resolution of, for example, about 1 degree, but is not limited thereto.

  The system 1 also includes a fuel tank 21, a low pressure pump 22, a pressure regulator 23, a low pressure fuel pipe 25, a low pressure delivery pipe 26, port injection valves 271 to 274, and a fuel pressure sensor 28.

  The fuel tank 21 stores gasoline as fuel. The low pressure pump 22 pressurizes the fuel and discharges it into the low pressure fuel pipe 25. The pressure regulator 23 regulates the fuel discharged into the low-pressure fuel pipe 25 to a preset supply pressure on the low-pressure side.

  The low-pressure fuel pipe 25 and the low-pressure delivery pipe 26 are an example of a low-pressure fuel passage that supplies the fuel discharged from the low-pressure pump 22 to the port injection valve group 27. The fuel pressurized to a predetermined pressure level by the low-pressure pump 22 and adjusted to the low-pressure supply pressure by the pressure regulator 23 is introduced into the low-pressure delivery pipe 26 via the low-pressure fuel pipe 25.

  The port injection valve group 27 is connected to the low pressure delivery pipe 26 and injects fuel into the intake ports 13 corresponding to the cylinder groups 11 respectively. The fuel pressure sensor 28 detects the fuel pressure value in the low pressure delivery pipe 26, as will be described in detail later. The detected value of the fuel pressure sensor 28 is acquired by the ECU 41 at regular sampling time intervals.

  The system 1 includes a high pressure pump 31, a high pressure fuel pipe 35, a high pressure delivery pipe 36, in-cylinder injection valves 371 to 374, and a fuel pressure sensor 38.

  The high pressure pump 31 sucks fuel from the branch pipe 25 a branched from the low pressure fuel pipe 25 and pressurizes the fuel to a high pressure level higher than the supply pressure level from the low pressure pump 22. The branch pipe 25a is provided with a pulsation damper 29 for suppressing fuel pressure pulsation in the branch pipe 25a.

  Specifically, the high-pressure pump 31 includes a pump housing 31h, a plunger 31p that can slide in the pump housing 31h, and a pressurizing chamber 31a defined between the pump housing 31h and the plunger 31p. The volume of the pressurizing chamber 31a changes according to the displacement of the plunger 31p. Fuel pressurized by the low-pressure pump 22 is introduced into the pressurizing chamber 31a via the branch pipe 25a in a state where an electromagnetic valve 32 described later is opened. The fuel in the pressurizing chamber 31a is pressurized to a high pressure by the plunger 31p and discharged into the high-pressure fuel pipe 35.

  A cam CP for driving the plunger 31p is mounted on the cam shaft 15 of the engine 10. The shape of the cam CP is a substantially square with rounded corners. The high-pressure pump 31 includes a follower lifter 31f that is moved up and down by the cam CP, and a spring 31g that biases the follower lifter 31f toward the cam CP. The plunger 31p is interlocked with the follower lifter 31f, and the plunger 31p is also moved up and down together with the follower lifter 31f. The camshaft 15 is interlocked with the crankshaft 14 via a chain or a belt. The cam shaft 15 and the cam CP are driven at a rotational speed that is ½ of the rotational speed of the crankshaft 14.

  An electromagnetic valve 32 is provided at the fuel inlet of the pressurizing chamber 31a of the high-pressure pump 31. The electromagnetic valve 32 includes a valve body 32v, a coil 32c that drives the valve body 32v, and a spring 32k that always biases the valve body 32v in the opening direction. Energization of the coil 32 c is controlled by the ECU 41 via the driver circuit 42. When the coil 32c is energized, the valve body 32v shuts off the branch pipe 25a and the pressurizing chamber 31a of the low-pressure fuel pipe 25 against the urging force of the spring 32k. In a state where the coil 32c is not energized, the valve body 32v is kept open by the urging force of the spring 32k.

  A check valve 34 with a spring is provided in the high pressure fuel pipe 35 between the high pressure delivery pipe 36 and the pressurizing chamber 31a. The check valve 34 opens when the fuel pressure on the pressurizing chamber 31a side is higher than the fuel pressure on the high pressure delivery pipe 36 side by a predetermined amount.

  In the suction stroke of the high-pressure pump 31, the solenoid valve 32 is opened and the plunger 31p is lowered, and fuel is charged into the pressurizing chamber 31a from the branch pipe 25a of the low-pressure fuel pipe 25. In the pressurizing stroke, the electromagnetic valve 32 is closed and the volume of the pressurizing chamber 31a is reduced as the plunger 31p is raised, and the fuel in the pressurizing chamber 31a is pressurized. In the discharge stroke, the check valve 34 is opened when the force of the fuel pressure in the pressurizing chamber 31a becomes larger than the biasing force of the spring of the check valve 34, and the pressurized fuel is supplied to the high pressure fuel pipe 35 and the high pressure delivery pipe 36. Supplied to. As described above, the raising / lowering of the plunger 31p is realized by the rotation of the cam CP. Since the cam CP is linked to the crankshaft 14 via the camshaft 15, the high-pressure pump 31 is driven to be linked to the crankshaft 14. The

  In this case, the solenoid valve 32 is opened when not energized, but is not limited thereto. For example, the solenoid valve 32 may be in a closed state with no energization, with the urging directions of the coil 32c and the spring 32k reversed. In this case, the coil 32c is energized during the fuel intake stroke, and is de-energized during the pressurization and discharge strokes.

  High pressure fuel pressurized by the high pressure pump 31 is stored in the high pressure delivery pipe 36 via the high pressure fuel pipe 35. The high-pressure fuel pipe 35 and the high-pressure delivery pipe 36 are an example of a high-pressure fuel passage that supplies high-pressure fuel from the high-pressure pump 31 to the cylinder injection valves 371 to 374.

  The in-cylinder injection valve group 37 directly injects high-pressure fuel from the high-pressure delivery pipe 36 into each of the cylinders 111 to 114 in a predetermined order. The fuel pressure sensor 38 detects the fuel pressure in the high-pressure delivery pipe 36, and the detected value of the fuel pressure sensor 38 is acquired by the ECU 41 at regular sampling time intervals.

  The ECU 41 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory). The ECU 41 executes control for predicting a fuel pressure value, which will be described later, based on information from the sensor, information stored in the ROM in advance, and the like according to a control program stored in the ROM in advance. This control is functionally realized by a CPU, ROM, and RAM, a central fuel pressure calculation unit, an amplitude calculation unit, a phase calculation unit, a storage unit, a prediction unit, an acquisition unit, a candidate calculation unit, a temporary fuel pressure value calculation unit, It is executed by the candidate specifying unit, the pulsation determining unit, and the injection start determining unit. Details will be described later.

  The ECU 41 calculates the required fuel injection amount in accordance with the operating state of the engine 10 and the acceleration request. Specifically, the ECU 41 calculates each injection period of the port injection valve group 27 corresponding to the required injection amount, and each of the port injection valve groups 27 is calculated in a predetermined order at predetermined crank angle intervals. Fuel injection is executed only for a period. The same applies to the in-cylinder injection valve group 37.

  The fuel injection amount of each of these fuel injection valves is proportional to the valve opening period. The valve opening period is proportional to the energization period of the electromagnetic coil of the fuel injection valve. Therefore, the ECU 41 calculates each energization period of the port injection valve group 27 according to the required injection amount based on the detection value of the fuel pressure sensor 28. Similarly, the ECU 41 calculates each energization period of the in-cylinder injection valve group 37 according to the required injection amount based on the detection value of the fuel pressure sensor 38. The ECU 41 issues a command to the driver circuit 42 according to the calculated energization period. The driver circuit 42 energizes each of the port injection valve group 27 and the in-cylinder injection valve group 37 for the calculated energization period in accordance with a command from the ECU 41. In this way, the fuel injection amount of each fuel injection valve is controlled.

  Next, fuel pressure pulsation caused by the high pressure pump 31 will be described. FIG. 2 is a waveform diagram of fuel pressure. The vertical axis represents the fuel pressure, and the horizontal axis represents the engine speed. As shown in FIG. 2, the engine speed range includes a pulsation increasing region in which the amplitude of the fuel pressure pulsation increases in the low pressure fuel pipe 25 and the low pressure delivery pipe 26 as compared with other regions. The pulsation increasing range is, for example, an engine speed of 1000 to 1200 rpm, but is not limited thereto.

  The reason why the amplitude of the fuel pressure pulsation increases as described above can be considered as follows. In a predetermined region of the engine speed, the in-cylinder injection valve group 37 is not used and fuel injection is performed by the port injection valve group 27. In the meantime, the in-cylinder injection valve group 37 is not used, so that the solenoid valve 32 is maintained in the open state, and the plunger 31p repeatedly moves up and down by the power of the engine 10. For this reason, the suction and discharge of fuel are repeated between the low-pressure fuel pipe 25 and the pressurizing chamber 31a, thereby increasing the amplitude of pulsation and propagating to the low-pressure delivery pipe 26. Further, when the frequency of the fuel pressure pulsation and the natural frequency of the pulsation damper 29 coincide with each other and resonate, the amplitude of the fuel pressure pulsation further increases.

  FIG. 3 is a graph showing an example of the waveform of the fuel pressure pulsation and the injection period by the port injection valve. The vertical axis represents fuel pressure, and the horizontal axis represents time. FIG. 3 shows a pulsation waveform in a state where the engine speed belongs to the above-described pulsation increasing region. Here, the fuel pressure value PA is an actual fuel pressure value during the port injection period, and the detection value PS is a detection value of the fuel pressure sensor 28 acquired by the ECU 41. Generally, based on the detected value PS acquired before the start of port injection, the injection period of port injection is calculated and the port injection amount is controlled. This is because the ECU 41 needs to complete the calculation of the port injection period based on the detected value of the fuel pressure sensor 28 before the port injection reaches the start timing. Further, the ECU 41 can acquire the detection value of the fuel pressure sensor 28 only at a constant sampling time interval. However, when the engine speed belongs to the above-described pulsation increasing region, the fuel pressure fluctuates in a short time, so that the difference between the detected value PS and the fuel pressure value PA may increase as shown in FIG. For this reason, if the port injection period is calculated based on the detection value PS, the port injection amount may not be accurately controlled.

  Therefore, in this embodiment, the ECU 41 predicts a future fuel pressure value based on a pulsation model formula described later. In the present embodiment, the control target controlled based on the predicted fuel pressure value is the port injection amount, but is not limited thereto. Further, in the following description, unless otherwise specified, “detected value” means a detected value of the fuel pressure sensor 28.

  As described above, the ECU 41 acquires detection values at regular sampling time intervals. FIG. 4A is a graph showing a plurality of detection values acquired by the ECU 41. In FIG. 4A, the vertical axis represents the fuel pressure and the horizontal axis represents the crank angle. As shown in FIG. 4A, conventionally, only the fuel pressure value at the present time or in the past can be grasped. FIG. 4B is a graph showing the modeled fuel pressure pulsation described below. Similarly in FIG. 4B, the vertical axis represents the fuel pressure and the horizontal axis represents the crank angle. In the present embodiment, the future fuel pressure value can be predicted by modeling the fuel pressure pulsation as shown in FIG. 4B. Further, the fuel pressure value within the sampling time interval of the detected value can also be predicted.

ここで、θ［ｄｅｇ］はクランク角である。Ｐ（θ）［ｋＰａ］は、クランク角θに対応した燃圧値である。Ａ［ｋＰａ］は、脈動の振幅である。Ｂ［ｄｅｇ］は、脈動の初期位相である。ｃは、クランク角３６０度当たりの高圧ポンプ３１の燃料の吐出回数であり、換言すれば、クランク角３６０度当たりの脈動の振動数である。Ｐ_ｃ［ｋＰａ］は、脈動の中心燃圧値である。中心燃圧値Ｐ_ｃは、振幅Ａの中心値である。初期位相Ｂは、クランク角θがゼロ度に最も近くて振幅Ａが極大値をとる際のクランク角である。尚、クランク角θがゼロ度になる時は、気筒１１１のピストンが圧縮行程の上死点に位置する時である。 Specifically, the ECU 41 predicts the fuel pressure value based on the following model equation (2).

Here, θ [deg] is a crank angle. P (θ) [kPa] is a fuel pressure value corresponding to the crank angle θ. A [kPa] is the amplitude of pulsation. B [deg] is an initial phase of pulsation. c is the number of times the fuel is discharged from the high-pressure pump 31 per 360 ° crank angle, in other words, the pulsation frequency per 360 ° crank angle. P _c [kPa] is the central fuel pressure value of pulsation. The central fuel pressure value _Pc is the central value of the amplitude A. The initial phase B is a crank angle when the crank angle θ is closest to zero degrees and the amplitude A takes a maximum value. The crank angle θ is zero degrees when the piston of the cylinder 111 is located at the top dead center of the compression stroke.

The reason why the fuel pressure pulsation can be modeled by an equation using a trigonometric function is that the pulsation periodically changes due to the rotation of the cam CP of the high-pressure pump 31 as described above. Here, the ECU 41 calculates the center fuel pressure value P _c , the amplitude A, and the initial phase B, which will be described in detail later.

  The number of discharges c is stored in advance in the ROM of the ECU 41. Here, the number of discharges c is determined by the shape of the cam CP of the high-pressure pump 31. In this embodiment, the shape of the cam CP is a substantially square with rounded corners. Therefore, the cam CP rotates 180 degrees per 360 degrees of crank angle and discharges fuel twice. Therefore, in the present embodiment, the number of ejections c is 2. In the case of a substantially equilateral triangular cam with rounded corners, the cam rotates 180 degrees per 360 degrees of crank angle, and the number of discharges c is 1.5. In the case of a substantially elliptic cam, the number of discharges c per crank angle of 360 degrees is 1.

  Next, port injection control that is executed based on the fuel pressure value predicted by the model equation (2) will be described. FIG. 5 is a flowchart illustrating an example of port injection control. In the following description, “previous value”, “previous value”, and “current value” mean the detected values of the fuel pressure sensor 28 acquired by the ECU 41 the previous time, previous time, and this time, respectively. The “next value” means a detection value of the fuel pressure sensor 28 that the ECU 41 is scheduled to acquire next time.

  The ECU 41 determines whether or not the engine speed [rpm] calculated based on the crank angle sensor 14a is less than a predetermined threshold (step S1). Here, the threshold is defined as the upper limit of the engine speed at which port injection can be requested, and is set to a value larger than the engine speed in the pulsation increasing region. If the determination is negative, the port injection is not executed and this control is terminated.

If the determination in step S1 is affirmative, the ECU 41 acquires a predetermined number of detection values and stores them in the RAM (step S2). When a predetermined number of detection values already acquired are stored in the RAM, the detection values are updated in order from the newly acquired detection values. The predetermined number is the number of detected values necessary for calculating the central fuel pressure value _Pc , as will be described in detail later. Further, the process of step S2 is also executed for calculating the amplitude A and the initial phase B. The detection values acquired and stored in step S2 include at least two previous detection values, the previous value and the current value. The process of step S2 is an example of a process executed by an acquisition unit that acquires the detection value of the fuel pressure sensor 28 at a constant sampling time interval.

Next, the ECU 41 executes a central fuel pressure calculation process for calculating the pulsation central fuel pressure value _Pc based on at least the previous value and the current value stored in the RAM in step S2 (step S3). Therefore, even when the engine speed does not belong to the above-described pulsation increase region, the central fuel pressure calculation process is executed. This is because when the engine speed belongs to the pulsation increasing region, the fuel pressure value can be immediately predicted by the model equation (2). The process of step S3 is an example of a process executed by a central fuel pressure calculation unit that calculates a central fuel pressure value _Pc of pulsation. Details will be described later.

  Next, the ECU 41 determines whether or not the engine speed belongs to the above-described pulsation increasing region (step S4). The pulsation increasing region is calculated in advance by experiments and stored in the ROM. The process of step S4 is an example of a process executed by a pulsation determining unit that determines whether or not the engine speed calculated based on the output of the crank angle sensor 14a belongs to the above-described pulsation increasing region.

  If the determination in step S4 is negative, the ECU 41 executes normal port injection processing on the assumption that the fuel pressure value does not vary greatly (step S15). The normal port injection process calculates the injection period based on the detection value acquired by the ECU 41 immediately before each injection start timing of the port injection valve group 27 without performing the above-described prediction of the fuel pressure value, and performs port injection. Is a process of executing. This is because when the amplitude of the pulsation is relatively small, the difference between the detected value immediately before the injection start timing and the fuel pressure value during port injection is small, and the port injection amount can be appropriately controlled. After executing the normal port injection process, the processes after step S1 are executed again.

  If the determination in step S4 is affirmative, the ECU 41 stores the crank angle at the time of acquisition of the current value acquired in step S2 in the RAM based on the output value from the crank angle sensor 14a (step S5). When the engine speed belongs to the pulsation increasing region, the crank angle at the time of acquiring the current value is updated every time a new detection value is acquired. Next, the ECU 41 converts the sampling time interval stored in advance in the ROM into a crank angle based on the latest engine speed (step S6). The reason for considering the engine speed is that the sampling speed interval is constant, but the engine speed fluctuates. Next, the ECU 41 adds the crank angle corresponding to the sampling time interval to the crank angle at the time of acquisition of the current value, and calculates the crank angle at the time of acquisition of the next value (step S7).

  Next, the ECU 41 determines whether any injection start timing of the port injection valve group 27 is before acquisition of the next value based on the crank angle at the time when the next value is scheduled to be acquired (step S8). ). Here, since the injection start timing of any one of the port injection valve groups 27 is associated with the crank angle in advance and stored in the RAM, the above determination can be made. In addition, the process of step S5-S7 is a process required in order to perform the determination process of step S8. The processing of steps S5 to S8 is executed by an injection start determination unit that determines whether the fuel injection start timing of any of the port injection valve groups 27 is after acquisition of the current value and before acquisition of the next value. It is an example of the process to perform. If a negative determination is made in step S8, the processes after step S1 are executed again, and the central fuel pressure calculation process is continued.

  If the determination in step S8 is affirmative, the ECU 41 executes an amplitude calculation process for calculating the amplitude A of the pulsation based on at least the previous value and the current value stored in the RAM in step S2 (step S9). Details will be described later. Next, the ECU 41 executes a phase calculation process for calculating the initial phase B based on at least one of the previous value and the current value stored in the RAM in step S2 (step S10). Details will be described later. As described above, the terms of the model formula (2) are calculated by the processes of steps S3, S9, and S10.

Next, the ECU 41 predicts a fuel pressure value corresponding to the crank angle during the injection period of the port injection valve scheduled for the next injection (step S11). Specifically, the values calculated by the processes of steps S3, S9, and S10 are substituted into each term of the model equation (2), and the fuel pressure value at a predetermined crank angle is predicted. Here, as described above, the value of each term of the model equation (2) is calculated based on the detection value of the fuel pressure sensor that reflects the actual fuel pressure value. For this reason, the value of each term of the model equation (2) is calculated with high accuracy, and the future fuel pressure value can be accurately predicted based on the model equation (2). The process of step S11 is performed based on the fuel pressure value P (corresponding to the crank angle θ [deg] based on the central fuel pressure value P _c , the amplitude A, the initial phase B, the number of discharges c, and the model equation (2). It is an example of the process which the estimation part which estimates (theta)) [kPa] performs. The process of step S11 is a predicting unit that predicts the fuel pressure value P (θ) when the fuel injection start timing of the port injector is after acquisition of the current value and before acquisition of the next detection value. It is an example of the process performed.

Ｑ［ｍＬ］は要求噴射量である。Ｑ_ＩＮＪ［ｍＬ／ｍｉｎ］は、ポート噴射弁２７１〜２７４の各公称流量である。Ｐ_０［ｋＰａ］は、ポート噴射弁２７１〜２７４の各公称流量に対応した検査圧力である。Ｑ_ＩＮＪ及びＰ_０は、予め実験により算出されてＲＯＭに記憶されている。Ｐ（θ）［ｋＰａ］は、モデル式（２）により予測された燃圧値である。 The ECU 41 calculates the energization period of the port injection valve scheduled for the next injection based on the predicted fuel pressure value (step S12). Specifically, the energization period of the port injection valve scheduled for the next injection is calculated based on the predicted fuel pressure value so as to inject the fuel by the required injection amount set according to the operation request of the engine 10. Specifically, the energization period α [ms] is calculated by the following equation.

Q [mL] is a requested injection amount. Q _INJ [mL / min] is each nominal flow rate of the port injection valves 271 to 274. P ₀ [kPa] is an inspection pressure corresponding to each nominal flow rate of the port injection valves 271 to 274. Q _INJ and P ₀ are calculated in advance by experiments and stored in the ROM. P (θ) [kPa] is the fuel pressure value predicted by the model equation (2).

  Next, the ECU 41 determines whether or not the current crank angle has reached the start timing of the next port injection valve injection based on the detected value of the crank angle sensor 14a (step S13). If the determination in step S13 is negative, the process in step S13 is executed again. If the determination in step S13 is affirmative, the ECU 41 conducts port injection by energizing the port injection valve scheduled to be injected for the calculated energization period (step S14). The process of steps S12 to S14 is an example of a process executed by an injection control unit that controls the fuel injection amount of the port injection valve based on the predicted fuel pressure value P (θ).

  As described above, before reaching the injection start timing of the port injection valve scheduled to be injected, the fuel pressure value within the scheduled port injection execution period is predicted by the model equation (2), and based on the predicted fuel pressure value. The energization period of the port injection valve is calculated. For this reason, even when the amplitude of pulsation is large, the fuel injection amount of the port injection valve can be accurately controlled, and the air-fuel ratio can be accurately controlled. Further, since the amplitude A and the initial phase B are calculated and the fuel pressure value is predicted only when an affirmative determination is made in step S8, the fuel pressure value is predicted only when necessary, and an increase in the processing load of the ECU 41 is suppressed. Yes.

  The future fuel pressure value predicted by the model equation (2) may be used for control other than the fuel injection amount of the port injection valve. For example, the ignition timing of the spark plug may be controlled based on the predicted future fuel pressure value. In the case where a certain period is required until the calculation process for controlling the control target is completed after the detection value of the sensor is acquired, such a predicted future fuel pressure value is used instead of the detection value of the sensor. It is effective.

  Further, even if the engine speed does not belong to the above-described pulsation increasing region, the future fuel pressure value predicted by the model equation (2) may be predicted. For example, since the low-pressure pump 22 is feedback-controlled as will be described later, the fuel pressure slightly pulsates due to fluctuations in the rotation of the low-pressure pump 22 even if the engine speed is not included in the pulsation increasing region. Because there is also.

Next, a plurality of examples of the central fuel pressure calculation process will be described. The ECU 41 calculates the center fuel pressure value _Pc by executing any one of a plurality of examples of the center fuel pressure calculation process described below.

First, a first example of the central fuel pressure calculation process will be described. In the first example, the center fuel pressure value _Pc is calculated based on the previous time value, the previous value, and the current value acquired in step S2 and the crank angle corresponding to the sampling time interval. FIG. 6 is a flowchart showing a first example of the central fuel pressure calculation process. The ECU 41 converts the sampling time interval into a crank angle (step S21). For step S21, the same technique as in step S6 is used.

ここで、係数Ｐ´＝（Ｐ_２−Ｐ_１）／（Ｐ_３−Ｐ_１）と定義すると、係数Ｐ´は式（４）〜（７）により、以下の式で表される。

上記式（８）の係数ａを解くと以下の式で表される。

Here, assuming that the crank angle t [deg] corresponding to the sampling time interval is a coefficient a, the previous value P ₁ [kPa], the previous value P ₂ [kPa], and the current value P ₃ [kPa] are as follows: It is expressed by the following formula.
(Equation 4)
P ₁ = Asin (a) + P _c (4)
(Equation 5)
P ₂ = Asin (a + t) + P _c (5)
(Equation 6)
P ₃ = Asin (a + 2t) + P _c (6)
When the amplitude A is eliminated using the above equations (4) and (5), the central fuel pressure value _Pc is expressed by the following equation.

Here, when the coefficient P ′ = (P ₂ −P ₁ ) / (P ₃ −P ₁ ) is defined, the coefficient P ′ is expressed by the following expressions using Expressions (4) to (7).

When the coefficient a in the above equation (8) is solved, it is expressed by the following equation.

Accordingly, the ECU 41 calculates the central fuel pressure value _Pc by substituting the equation (9) into the equation (7) (step S22). As described above, the central fuel pressure value P _{c in} a short period of time is determined by the previous three detected values P ₁ , the previous value P ₂ , and the current value P _3, and the crank angle t corresponding to the sampling time interval. Can be calculated.

The reason why the coefficient a is expressed using the coefficient P ′ in the equation (9) is that the coefficient P ′ is based only on the previous value P ₁ , the previous value P ₂ , and the current value P ₃ , which are detection values of the fuel pressure sensor 28. This is because the coefficient a is easily calculated by using the coefficient P ′. Thereby, calculation of the center fuel pressure value _Pc is facilitated, and an increase in the processing load of the ECU 41 and an increase in the calculation period are also suppressed.

In the first example of the central fuel pressure calculation process, since the central fuel pressure value _Pc is calculated based on the latest three detected values as described above, the predetermined number in step S2 may be three or more.

Next, a second example of the central fuel pressure calculation process will be described. In the second example, the smoothed value of the detected value is calculated as the central fuel pressure value _Pc . FIG. 7 is a flowchart showing a second example of the central fuel pressure calculation process.

なまし値Ｐ´_ｎは、今回算出されるなまし値である。なまし値Ｐ´_ｎ−１［ｋＰａ］は、前回算出されたなまし値である。Ｐ_ｎ［ｋＰａ］は、今回値である。ｋ［‐］は、なまし係数である。なまし係数ｋの初期値は、１６である。ここで、なまし値Ｐ´_ｎの算出には、今回値Ｐ_ｎとなまし値Ｐ´_ｎ−１とに基づいて算出されるが、なまし値Ｐ´_ｎ−１は前回算出されたなまし値であって前回値Ｐ_ｎ−１に基づいて算出されたものである。このため、なまし値Ｐ´_ｎを算出するためには、少なくとも直近の２つの検出値である前回値及び今回値が必要である。 The ECU 41 calculates the smoothed value P ′ _n [kPa] as the central fuel pressure value P _c by the following equation based on the predetermined number of detection values acquired in step S2 (step S31).

The annealing value P ′ _n is the annealing value calculated this time. The annealing value P ′ _n−1 [kPa] is the annealing value calculated last time. P _n [kPa] is the current value. k [-] is an annealing coefficient. The initial value of the annealing coefficient k is 16. Here, I the calculation of the smoothed value _P'n, but is calculated based on the Shi value _P'n-1 raw between the present value P _n, smoothed value _P'n-1 is calculated the last time It is a preferred value and is calculated based on the previous value P _n−1 . For this reason, in order to calculate the annealing value P ′ _n , at least the last two detected values, the previous value and the current value, are necessary.

Figure 8 is a graph showing the moderation and the obtained detected value value _P'n. A black circle in FIG. 8 indicates a detection value acquired by the ECU 41. FIG. 8 shows the annealing value P ′ _n when the annealing coefficient k is 16, and the annealing value P ′ _n is the center fuel pressure value P _c and is calculated every time the detected value is acquired. The annealing value P ′ _n is shown continuously. The calculation of the annealing value P ′ _n is performed every time the detection value is acquired during the occurrence of pulsation. Since the central fuel pressure value P _c is calculated based on a plurality of detected values as, accurately center the fuel pressure value P _c is calculated. It should be noted that the annealing process is first executed with the current value, and if the previously calculated annealing value does not exist, the previous value is calculated as the annealing value P ′ _n−1 .

Next, the ECU 41 determines whether or not the absolute value of the difference between the central fuel pressure value P _c calculated by the above-described annealing process and the target fuel pressure value P ₀ [kPa] is equal to or greater than a predetermined value E [kPa]. Determination is made (step S32). Here, the target fuel pressure value P _0, a target value of the fuel pressure in the low-pressure delivery pipe 26, for example, the fuel is set to a low pressure so as not to boiling under reduced pressure. ECU41 is smoothed value of the detected values such that it converges to the target fuel pressure value P _0, the feedback control of the voltage applied to the low-pressure pump 22. Therefore, the pulsation is mainly caused by the high pressure pump 31, but the central fuel pressure value P _c of the pulsation is mainly caused by the low pressure pump 22.

If the determination in step S32 is affirmative, that is, if E ≧ | P _c −P ₀ | is satisfied, the ECU 41 sets the smoothing coefficient k to 16 (step S33). In this case, it means that the difference between the central fuel pressure value P _c and the target fuel pressure value P ₀ is small, by setting a large averaging coefficient k, can be calculated accurately center the fuel pressure value P _c. As a result, as long as an affirmative determination is made in step S32, the smoothing process is performed with the smoothing coefficient k being 16 each time a new detection value is acquired.

On the other hand, if the determination in step S32 is negative, that is, if E ≧ | P _c −P ₀ | is not satisfied, the ECU 41 sets the smoothing coefficient k to 8 (step S34). Here, the case where E ≧ | P _c −P ₀ | is not satisfied means that the difference between the central fuel pressure value P _c and the target fuel pressure value P ₀ is large, and the fuel pressure in the low pressure delivery pipe 26 is the target fuel pressure value. This means that the low pressure pump 22 is controlled so as to converge to P ₀ . Therefore, in this case, changes transiently as the fuel pressure in the low-pressure delivery pipe 26, that is, actual center fuel pressure value converges to the target fuel pressure value P _0. In such a case, when the annealing coefficient k is small, the actual change in the fuel pressure is more easily reflected in the annealing value P ′ _n than when it is large. Therefore, by setting the annealing coefficient k to 8, the center fuel pressure value _Pc can be calculated following such a transient change in fuel pressure.

In the second example of the central fuel pressure calculation process, since the central fuel pressure value _Pc is calculated based on the two most recent detection values as described above, the predetermined number in step S2 may be two or more.

The minimum value of the smoothing coefficient k is not limited thereto but is 6, for example, if the difference between the target fuel pressure value P ₀ and the center fuel pressure value Pc is larger, as compared with smaller, averaging coefficient k is Small is enough. Therefore, as the center fuel pressure value P _c and the difference calculated between the target fuel pressure value P ₀ is large, may be set averaging coefficient k is small. The minimum value of the annealing coefficient k is set in consideration of, for example, the sampling period, the amplitude of the fuel pressure at the time of the resonance occurrence measured by experiments, the maximum change speed of the central fuel pressure value, and the like. The annealing value of the fuel pressure used for controlling the low pressure pump 22 may be the same as or different from the annealing value used for calculating the center fuel pressure value _Pc .

Next, a third example of the central fuel pressure calculation process will be described. In the third example, the ECU 41 calculates the average value of the acquired detection values as the central fuel pressure value _Pc . FIG. 9 is a flowchart showing a third example of the central fuel pressure calculation process.

The ECU 41 calculates the engine speed based on the output signal from the crank angle sensor 14a (step S41), and calculates the number N of detection values used for calculating the center fuel pressure value _Pc based on the engine speed. (Step S42). Specifically, the engine speed Ne [rpm], the sampling time interval Ts [ms] for obtaining the detection value, the number of discharges c, and the number of samples according to the equation N = {60000 / (Ne · Ts · c)} + 1 N is calculated. In addition, the value after the decimal point of the calculated sample number N is rounded off. Here, the sampling time interval Ts is stored in advance in the ROM of the ECU 41.

FIG. 10A is a graph showing the relationship between the number N of detected values used for calculating the central fuel pressure value _Pc and the engine speed. The vertical axis indicates the number of samples N, and the horizontal axis indicates the engine speed.

  The calculated sample number N corresponds to the number of detection values acquired during a period of approximately one cycle of pulsation. Here, the pulsation period becomes shorter as the engine speed increases, that is, as the rotational speed of the cam CP of the high-pressure pump 31 increases. On the other hand, the sampling time interval at which the detection value is acquired is constant as described above. For this reason, the number of detection values acquired in the period for one cycle of pulsation decreases, so that engine speed increases. Therefore, as shown in FIG. 10A, the sample number N decreases as the engine speed increases.

Next, the ECU 41 calculates the average value of the calculated detection values for the number of samples N as the central fuel pressure value _Pc (step S43). The detection value extracted as a sample is a detection value continuously detected before this value.

  FIG. 10B and FIG. 10C are graphs showing the fuel pressure pulsation and the number of samples. FIG. 10B shows a case where the engine speed is small, and FIG. 10C shows a case where the engine speed is large. FIG. 10B shows a case where the number of samples N is 9, and FIG. 10C shows a case where the number of samples N is 5.

Each time a detection value of a predetermined number of samples N is acquired, the ECU 41 calculates the average value of the detection values as the central fuel pressure value _Pc . Therefore, the ECU 41 can accurately calculate the center fuel pressure value _{Pc in accordance} with the change in the engine speed. Further, since the central fuel pressure value Pc can be calculated by a simple method, an increase in the processing load of the ECU 41 and a prolonged calculation process are also suppressed.

  Note that the ECU 41 may determine the number of samples N based on a map as shown in FIG. 10A stored in advance in the ROM, instead of determining based on the above-described calculation formula.

In addition, when the sampling time interval Ts is relatively long and the engine speed Ne is relatively high, or when the number of discharges c is relatively large, the number of samples may be 2. For this reason, in the third example of the central fuel pressure calculation process, since the central fuel pressure value _Pc can be calculated based on the two most recent detection values, the predetermined number in step S2 is at least two or more.

As described above, three examples of the central fuel pressure calculation process have been described. In any case, the central fuel pressure value _Pc is calculated based on at least the previous value and the current value.

式（１１）及び（１２）が中心燃圧値Ｐ_ｃを含んでいない理由は、中心燃圧値Ｐ_ｃをゼロとみなしても振幅Ａや初期位相Ｂの算出結果には影響がないからである。 Next, the amplitude calculation process will be described. Assuming that the respective crank angles at the time when the current value P _n and the previous value P _n−1 are acquired are θ _n and θ _n−1 , they can be expressed by the following equations (11) and (12).

Why the formula (11) and (12) does not contain a central fuel pressure value P _c is also regarded a central fuel pressure value P _c is zero because there is no effect on the calculation result of the amplitude A and the initial phase B.

このように振幅Ａは、中心燃圧値Ｐｃを用いずに、今回値Ｐ_ｎ及び前回値Ｐ_ｎ−１と、サンプリング時間間隔に対応するクランク角θ_ＡＤとにより、簡易な式で表される。このように簡易な式（１３）に基づいて振幅Ａが算出できるため、ＥＣＵ４１の処理負荷の増大や算出処理の期間の長期化が抑制される。 Here, the crank angle θ _AD [deg] corresponding to the sampling time interval can be expressed as θ _AD = θ _n −θ _n−1 . Therefore, the amplitude A can be expressed as follows based on the equations (11) and (12).

Thus the amplitude A is without a central fuel pressure value Pc, the current value P _n and the previous value P _n-1, by the crank angle theta _AD corresponding to the sampling time interval is expressed by a simple formula. Thus, since the amplitude A can be calculated based on the simple expression (13), an increase in the processing load of the ECU 41 and an increase in the period of the calculation process are suppressed.

ここで、Ｂ^＋＝θ_ｎ＋｛ｃｏｓ^−１（Ｐ_ｎ／Ａ）×（１／ｃ）｝、Ｂ⁻＝θ_ｎ−｛ｃｏｓ^−１（Ｐ_ｎ／Ａ）×（１／ｃ）｝とする。このように、候補Ｂ^＋及びＢ⁻は、今回値Ｐ_ｎが取得された時点でのクランク角θ_ｎ、今回値Ｐ_ｎ、振幅Ａ、及び吐出回数ｃにより簡易に表すことができる。尚、式（１４）において、クランク角θ_ｎ及び今回値Ｐ_ｎの代わりに、クランク角θ_ｎ−１及び前回値Ｐ_ｎ−１を用いてもよい。この場合も、理論上は初期位相Ｂの値は同じ値になる。ステップＳ５１は、前回値Ｐ_ｎ−１及び今回値Ｐ_ｎの一方と、前回値Ｐ_ｎ−１及び今回値Ｐ_ｎの一方が取得された時点でのクランク角と、吐出回数ｃと、算出された振幅Ａとに基づいて、初期位相Ｂの２つの候補Ｂ^＋及びＢ⁻を算出する候補算出部が実行する処理の一例である。 Next, the phase calculation process will be described. FIG. 11 is a flowchart illustrating an example of the phase calculation process. The ECU 41 calculates two candidates B ⁺ and B ⁻ for the initial phase B based on the following formula (14) calculated based on the formula (11) (step S51).

Here, B ⁺ = θ _n + {cos ⁻¹ (P _n / A) × (1 / c)}, B ⁻ = θ _n − {cos ⁻¹ (P _n / A) × (1 / c)} And In this way, the candidates B ⁺ and B ⁻ can be simply expressed by the crank angle θ _n , the current value P _n , the amplitude A, and the number of discharges c when the current value P _n is acquired. In the equation (14), the crank angle θ _n−1 and the previous value P _n−1 may be used instead of the crank angle θ _n and the current value P _n . Also in this case, the value of the initial phase B is theoretically the same value. Step S51 is the one of the previous value P _n-1 and the current value P _n, and the crank angle at the time that one is acquired in previous value P _n-1 and the current value P _n, and discharge count c, is calculated 8 is an example of processing executed by a candidate calculation unit that calculates two candidates B ⁺ and B ⁻ of the initial phase B based on the amplitude A.

Here, in order to specify the true solution of the initial phase B, the temporary fuel pressure value obtained by substituting one of the two candidates B ⁺ and B ⁻ into the model equation (2), the actual detection value, It is possible to specify based on whether or not. For example, the temporary fuel pressure value P _n−1 ⁺ is calculated by substituting the calculated center fuel pressure value P _c, amplitude A, crank angle θ _n−1 and candidate B ⁺ into the model equation (2). If the temporary fuel pressure value P _n-1 ⁺ calculated based on this candidate B ⁺ matches the previous value P _n-1 corresponding to the temporary fuel pressure value P _n-1 ⁺ , the candidate B ⁺ is a true solution. , and the in the case of disagreement candidate B ^- it is the true solution. However, even if the true solution of the initial phase B is the candidate B ⁺ due to the calculation error of the central fuel pressure value _Pc , the temporary fuel pressure value _Pn-1 ⁺ and the previous value _Pn-1 are as described above. There is a possibility that they do not match, and there is a possibility that the initial phase B is calculated with an incorrect value.

Therefore, the ECU 41 calculates the temporary fuel pressure value P _n-1 ⁺ by substituting the calculated central fuel pressure value P _c, amplitude A, crank angle θ _n−1, and candidate B ⁺ into the model equation (2). (Step S52).

式（１５）の左辺にある、Ａｃｏｓ｛ｃ（θ_ｎ−１−Ｂ^＋）｝＋Ｐ_ｃは、上述した候補Ｂ^＋に基づいて算出された仮燃圧値Ｐ_ｎ−１ ^＋を意味する。所定値ε［ｋＰａ］は、初期位相Ｂの真の解が候補Ｂ^＋である場合における、仮燃圧値Ｐ_ｎ−１ ^＋と前回値Ｐ_ｎ−１との取り得る最大の誤差よりも若干大きい値であり、予め実験により算出されＲＯＭに記憶されている。式（１５）が成立する場合には、ＥＣＵ４１は候補Ｂ^＋が初期位相Ｂとして特定され（ステップＳ５４）、不成立の場合には候補Ｂ⁻が初期位相Ｂとして特定される（ステップＳ５５）。所定値εは、予めＲＯＭに記憶されているため、ＥＣＵ４１の処理負荷の増大が抑制されている。 Next, the ECU 41 determines whether or not the absolute value of the difference between the previous value P _n−1 and the above _- described temporary fuel pressure value P _n−1 ⁺ is less than a predetermined value (step S53). Specifically, the determination is made based on the following equation (15).

A cos {c (θ _n−1 −B ⁺ )} + P _{c on} the left side of the equation (15) means the temporary fuel pressure value P _n−1 ⁺ calculated based on the candidate B ⁺ described above. The predetermined value ε [kPa] is slightly larger than the maximum possible error between the temporary fuel pressure value P _n−1 ⁺ and the previous value P _n−1 when the true solution of the initial phase B is the candidate B ^+. This value is calculated in advance through experiments and stored in the ROM. If equation (15) is satisfied, ECU 41 candidate ^{B +} is identified as the initial phase B (step S54), the candidate B in the case of not satisfied ^- is identified as the initial phase B (step S55). Since the predetermined value ε is stored in the ROM in advance, an increase in the processing load on the ECU 41 is suppressed.

  As described above, since the final initial phase B can be calculated based on the simple inequality expression (15), the initial phase B can be accurately determined while suppressing an increase in the processing load of the ECU 41 and a prolonged calculation process. It can be calculated. Thereby, the future fuel pressure value finally predicted by the model equation (2) can be accurately predicted.

In step S52, the calculated amplitude A and central fuel pressure value P _c , one of the two candidates B ⁺ and B ⁻ , and the crank angle θ _n−1 at the time when the previous value P _n−1 was acquired are obtained. It is an example of the process which the temporary fuel pressure value calculation part which substitutes into Formula (2) and calculates temporary fuel pressure value _Pn-1 ⁺ performs.

Steps S53 to S55 are performed when the absolute value of the difference between the calculated temporary fuel pressure value P _n-1 ⁺ and the previous value P _n-1 corresponding to the temporary fuel pressure value P _n-1 ⁺ is less than a predetermined value. to identify candidate B ⁺ as an initial phase B, and if the absolute value of the difference exceeds the predetermined value, the candidate B ^- is an example of a candidate identifying unit that identifies the other as the initial phase B.

In equation (15), the direction of the inequality sign may be reversed. In this case, the candidate if the expression (15) is satisfied B ^- is, in the case of not satisfied candidate B ⁺ is identified as the initial phase B. In the equation (15), the candidate B ⁻ may be used instead of the candidate B ⁺ . In this case, the candidate B ⁻ is satisfied when the equation (15) is satisfied, and the candidate B ⁺ is satisfied when the equation (15) is not satisfied. Is identified as the initial phase B.

Moreover, ECU41 may perform the process of step S52 and S53 using Formula (16) instead of Formula (15).

A cos {c (θ _n−1 −B ⁻ )} + P _{c on} the right side of the equation (16) means a temporary fuel pressure value P _n−1 ⁻ calculated based on the candidate B ⁻ . Here, the right side of Expression (16) corresponds to a “predetermined value”, and the predetermined value is an absolute value of a difference between the previous value P _n−1 and the temporary fuel pressure value P _n−1 ⁻ .

When Expression (16) is satisfied, that is, when the left side is smaller than the right side, the previous value P _n−1 and the temporary fuel pressure value P _n−1 ⁺ are approximated, and the temporary fuel pressure value P _n−1 ⁺ Means close to the actual fuel pressure value. Therefore, in this case, the ECU 41 specifies the candidate B ⁺ as the initial phase B. When Formula (16) is not established, it means that the difference between the previous value P _n−1 and the temporary fuel pressure value P _n−1 ⁺ is large, and the previous value P _n−1 and the temporary fuel pressure value P _n−1 ⁻ Is an approximation. Thus, ECU 41 candidate B ^- identifies as the initial phase B.

In Expression (16), may be inequality orientation Conversely, this case, the candidate if the expression (16) is satisfied B ^- is a particular candidate B ⁺ is an initial phase B in the case of not satisfied Is done.

In the formula (15) and (16), in place of the previous value _{P n-1} and the crank angle theta _n-1, may be used current value _{P n} and the crank angle theta _n.

尚、位相遅れτは、予め実験により算出されておりＥＣＵ４１のＲＯＭに記憶されている。モデル式（１７）により燃圧値Ｐ（θ）を算出することにより、燃圧値をより精度よく予測できる。 Next, prediction of the fuel pressure value P (θ) in consideration of a delay from when the fuel pressure sensor 28 outputs an output signal corresponding to the detected value to when the ECU 41 acquires the detected value will be described. Here, the CPU of the ECU 41 acquires a signal after the output signal of the fuel pressure sensor 28 is filtered by the low-pass filter as a detection value. The low-pass filter removes a noise component included in the output signal of the fuel pressure sensor 28 by attenuating a high frequency component having a predetermined frequency or higher from the output signal of the fuel pressure sensor 28. Due to this filtering, there may be a delay from when the fuel pressure sensor 28 outputs an output signal corresponding to the detected value until the ECU 41 acquires the detected value. For example, as shown in FIG. 12, the model fuel pressure calculated by the above-described model equation (2) is delayed by a predetermined phase τ with respect to the actual fuel pressure. FIG. 12 is a graph showing a case where the model fuel pressure is delayed with respect to the actual fuel pressure. The ECU 41 may predict the fuel pressure value P (θ) based on the following model equation (17) taken as this phase τ [deg] instead of the model equation (2).

The phase delay τ is calculated in advance by experiments and stored in the ROM of the ECU 41. By calculating the fuel pressure value P (θ) from the model equation (17), the fuel pressure value can be predicted more accurately.

  Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. It can be changed.

1 Engine system (control device for internal combustion engine)
10 Engine (Internal combustion engine)
11 Cylinder Group 111-114 Cylinder 14 Crankshaft 14a Crank Angle Sensor 15 Camshaft 22 Low Pressure Pump 25 Low Pressure Fuel Pipe (Low Pressure Fuel Path)
26 Low pressure delivery pipe (low pressure fuel passage)
27 Port injection valve group 271 to 274 Port injection valve 28 Fuel pressure sensor 31 High pressure pump 35 High pressure fuel pipe (high pressure fuel passage)
36 High pressure delivery pipe (high pressure fuel passage)
37 In-cylinder injection valve group 371-374 In-cylinder injection valve 41 ECU (central fuel pressure calculation unit, amplitude calculation unit, phase calculation unit, storage unit, prediction unit, acquisition unit, candidate calculation unit, provisional fuel pressure value calculation unit, candidate specification Part, pulsation judgment part, injection start judgment part)
CP cam

Claims (1)

An in-cylinder injection valve that directly injects fuel into the cylinder of the internal combustion engine;
A port injection valve for injecting fuel into the intake port of the internal combustion engine;
A low pressure pump for pressurizing the fuel;
A low pressure fuel passage for supplying fuel pressurized by the low pressure pump to the plurality of port injection valves;
A high-pressure pump that is driven in conjunction with the internal combustion engine, further pressurizes the fuel supplied from the low-pressure fuel passage, and generates a pulsation of fuel pressure in the low-pressure fuel passage;
A high-pressure fuel passage branched from the low-pressure fuel passage and supplying fuel pressurized by the high-pressure pump to the plurality of in-cylinder injection valves;
A crank angle sensor for detecting a crank angle of the internal combustion engine;
A fuel pressure sensor for detecting a fuel pressure in the low pressure fuel passage;
An acquisition unit for acquiring the detection value of the fuel pressure sensor at a constant sampling time interval;
A central fuel pressure calculation unit that calculates a central fuel pressure value P _c [kPa] of the pulsation based on at least the previous value and the current value that are the detection values acquired last time and this time by the acquisition unit;
An amplitude calculation unit for calculating an amplitude A [kPa] of the pulsation based on at least the previous value and the current value;
A phase calculation unit that calculates an initial phase B [deg] of the pulsation based on at least one of the previous value and the current value;
A storage unit that stores the number of times c of fuel discharged by the high-pressure pump per 360 degrees of the crank angle;
Based on the calculated central fuel pressure value P _c , the amplitude A, the initial phase B, the number of discharges c, and the following equation (1), the fuel pressure value P ( θ) [kPa] predicting unit, and
The phase calculation unit includes the one of the previous value and the current value, the crank angle at the time when the one of the previous value and the current value is acquired, the number of discharges c, and the calculated amplitude. Based on A and the calculated central fuel pressure value P _c , the initial phase B is calculated,
The phase calculation unit includes:
Based on the one of the previous value and the current value, the crank angle at the time when the one of the previous value and the current value is acquired, the number of discharges c, and the calculated amplitude A. A candidate calculation unit for calculating two candidates for the initial phase B;
The calculated amplitude A and central fuel pressure value _Pc , one of the two candidates, the crank angle at which the one or the other of the previous value and the current value is acquired, and the number of discharges c Substituting into the equation (1) to calculate the temporary fuel pressure value,
When the absolute value of the difference between the calculated temporary fuel pressure value and the previous value or the current value corresponding to the temporary fuel pressure value is less than a predetermined value, the one of the candidates is specified as the initial phase B And a candidate specifying unit that specifies the other of the candidates as the initial phase B when the absolute value of the difference exceeds the predetermined value.

Images