JPH02277938A - Fuel feed control device of internal combustion engine - Google Patents

Abstract

PURPOSE:To improve the air-fuel ratio controllability by determining the estimated time to the preset crank angle position and the estimated change quantity of an engine load in the time added with the fuel feed cycle time, and calculating and selecting the corresponding correction quantity respectively. CONSTITUTION:A change quantity calculating means B calculates the change quantity of an engine load based on the signal from an operation state detecting means A synchronously with fuel feed quantity calculation. A correction quantity calculating means E calculates the time to the preset crank angle position calculated by the first estimated time calculating means C and the estimated change quantity of the engine load in the time added with the fuel feed cycle time calculated by the second estimated time calculating means F based on it to determine the first and second correction quantities corresponding to them. A correction quantity calculation implementation judging means D judges the implementation existence of a correction quantity calculating means E, if the implementation exists, the first correction quantity is selected by a selecting means G, if the implementation does not exists, the second correction quantity is selected. The estimated precision for the engine load change quantity is secured, and the air-fuel ratio controllability can be improved.

Description

[Detailed Description of the Invention] <Industrial Field of Application> The present invention relates to a fuel supply control device for an internal combustion engine, and in particular to a fuel supply control device for an internal combustion engine, which improves transient operation performance by highly accurate correction control of fuel supply amount during transient operation. Regarding equipment.

<Prior Art> The following types of fuel supply control devices for internal combustion engines are known. The intake air flow rate and intake pressure are detected as state quantities related to the intake air amount, and the basic fuel supply amount Tp is calculated based on these and the detected value of the engine rotation speed.

Then, this basic fuel supply amount Tp is adjusted to various correction coefficients C0EF, which are set based on operating conditions such as engine temperature, and a feedback correction coefficient LAMBDA, which is set based on the air-fuel ratio obtained through detection of the oxygen concentration in the exhaust gas. , calculate the final fuel injection amount Ti by correcting it with the correction numerator s based on the battery voltage, etc. Ti+-TpXCOE F
XLAMBDA+Ts), this fuel injection amount is intermittently supplied by a fuel injection valve or the like in synchronization with engine rotation.

By the way, in a device that calculates and sets the fuel supply amount and controls the fuel supply intermittently, there may be a detection delay of the air flow meter for detecting the intake air flow rate or a pressure sensor for detecting the intake pressure during transient operation, or the control may be delayed. This causes a delay in the calculation of the device.

For this reason, for example, during acceleration, the detected value at the time of determining the fuel supply amount is smaller than the actual intake air flow rate and intake pressure during the intake stroke, so the fuel supply amount is also smaller than the demand. This causes the air-fuel ratio to become overly lean, increasing the amount of nitrogen oxides NOx and hydrocarbons HC discharged, and causing acceleration shock and deterioration of acceleration response due to a delay in the response of the average effective pressure.

For this reason, the present applicant has proposed a method to perform fuel supply control that follows changes in the required fuel amount during transient operation by dealing with the detection delay and calculation delay, as well as taking into account the time difference between the time of detection and the time of the intake stroke. had previously proposed a fuel supply control system (Japanese Patent Application No. 62-269467).

In this system, the engine load amount is calculated from the throttle valve opening and the engine rotational speed in synchronization with the calculation of the normal fuel supply amount, which is performed every predetermined time (for example, every 10 seconds). The amount of change in engine load per unit time is calculated from the difference between the previous value and the current value.

Then, the time required from the calculation of the engine load change lid to the target crank angle position for fuel control, for example, intake BDC, is predicted and calculated, and the engine load up to the intake stroke is calculated based on the amount of change and the required time. A correction amount commensurate with the estimated amount of change is calculated, and the fuel supply amount is corrected and controlled based on the calculated correction amount.

<Problems to be Solved by the Invention> However, in this case, if the time required to reach the target crank angle position is the time required to reach the target crank angle position in the next cylinder to which fuel supply is started, then, for example, FIG. In the case shown in , the correction amount for the fuel supply amount to the #1 cylinder is the #2 cylinder, which is supplied with fuel before that.
Since correction amounts matched to the cylinders are used, there is a problem in that the desired correction control cannot be performed in the #1 cylinder and the air-fuel ratio ends up being the same.

In the example shown in FIG. 11, in a 4-cylinder engine, the crank angle sensor detects the
In synchronization with the reference angle signal REF outputted to the
Fuel supply is started for each cylinder, and the fuel supply amount used for normal fuel supply control in synchronization with the reference angle signal RE'F is updated every 10 m5. Then, in synchronization with the update calculation of the normal fuel supply amount every 10m5, the engine load change amount based on the throttle valve opening and engine rotation speed and the time from that time to the intake BDC of the nearest fuel supply start cylinder. The correction amount is calculated from this engine load change amount and time.

Therefore, when it is time to start fuel supply to the #2 cylinder, the fuel supply is controlled based on the normal fuel supply amount and the correction amount calculated at point A in the figure, and the correction amount is adjusted to the #2 cylinder. Since this corresponds to the time until the intake BDC of #2 cylinder, it is possible to perform a correction commensurate with the amount of load change from the time when the normal fuel supply amount is set until the intake BDC is reached. However, for the #1 cylinder to be supplied with fuel next, the fuel amount has not been updated since point A, so the normal fuel supply lid and correction amount set at point A will be used. As described above, since the correction amount corresponds to the engine load change up to the intake BDC of the #2 cylinder, the correction amount is insufficient in the #1 cylinder and the desired air-fuel ratio control cannot be performed.

That is, when the fuel supply amount and correction amount are calculated during the fuel supply timing (reference angle signal), the desired fuel correction control can be applied to the next cylinder to which fuel supply is to be started. However, at high revolutions), the calculation frequency at a constant crank angle decreases, and if the calculation of the correction amount is not updated between fuel supply timings, the desired correction control cannot be performed in the next cylinder to which fuel will be supplied. It's something that didn't exist. In particular, when the number of cylinders increases and the interval angle between the fuel supply start timings becomes smaller, problems such as those described in F will occur at lower engine speeds than when the number of cylinders is small.

In order to solve this problem, the time required to reach the target angle position can be set for each cylinder (Patent Application No. 1-2).
However, setting the required time for each cylinder has the problem of increasing memory capacity. In addition, it is possible to solve the above problem by shortening the calculation cycle of the fuel supply amount (correction amount) (for example, from 10m5 cycle to 5ms cycle), but it is possible to solve the above problem by shortening the calculation cycle. This increases the load on the CPU of the microcomputer, which is undesirable.

The present invention has been made in view of the above-mentioned problems, and is a fuel fuel system that predicts and sets the engine load change amount by taking into account the time difference up to the intake stroke, and sets the correction amount based on this predicted engine load change amount. It is an object of the present invention to enable prediction of an engine load change amount with high accuracy even at high engine speeds in supply correction control without increasing memory capacity or shortening the calculation cycle.

<Means for Solving the Problems> Therefore, in the present invention, as shown in FIG. a change amount calculation means that calculates the amount of change in the engine load based on the state quantity detected by the operating state detection means; and when the amount of change in the engine load is calculated by the change amount calculation means. a first predicted time calculation means that predicts and calculates the time required from a second prediction time setting means for setting a second 'F measurement waiting time by adding the fuel supply period time to the calculated first prediction time; Calculating a first correction amount of the fuel supply amount commensurate with the predicted change amount of the engine load in one prediction time, and predicting the engine load in the second prediction time based on the change amount of the engine load and the second prediction time. correction amount calculation means for calculating a second correction amount of the fuel supply amount commensurate with the amount of change;
A correction amount calculation execution determination means for determining whether or not the correction amount calculation means is executed during a fuel supply period synchronized with engine rotation, and when the correction amount calculation execution determination means determines whether or not the correction amount calculation means is to be executed. Select the first correction amount for 7
, a correction amount selection means for selecting the second correction amount when it is determined that the second correction amount is not to be executed, and a fuel supply correction means for correcting and controlling the fuel supply amount based on the selected correction amount.
.

In this configuration, the change amount calculation means is performed in synchronization with the calculation of the normal fuel supply amount performed at predetermined time intervals, and is based on the state quantity related to the engine load detected by the operating state detection means. Calculate the amount of change in engine load.

The first predicted time calculation means predicts and calculates the time required from when the change amount of the engine load is calculated by the change amount calculation means to a predetermined crank angle position, and calculates a predicted change in the engine load over this required time. Set to the first predicted time. Also, the second
The predicted time setting means sets a second predicted time by adding the fuel supply cycle time to the first predicted time.

The correction amount calculating means calculates a first correction amount of the fuel supply amount commensurate with the predicted change amount of the engine load in the first predicted time based on the engine load change cover and the first predicted time, and Based on the amount of change in the engine load and the second predicted time, a second correction amount of the fuel supply amount is calculated in accordance with the predicted amount of change in the engine load during the second predicted time. That is, the correction amount calculating means calculates a correction amount commensurate with the predicted amount of change in engine load in the time required to reach a predetermined crank angle position, and a time period longer than the time required to reach the predetermined crank angle position by the fuel supply cycle time. A correction amount commensurate with the predicted amount of change in engine load is calculated respectively.

The correction amount calculation execution determination means determines whether or not the correction amount calculation means is executed during a fuel supply period synchronized with engine rotation. The correction amount selection means selects the first correction amount when it is determined that the correction amount calculation means is to be executed, and when it is determined that the correction amount calculation means is not to be executed, the correction amount selection means selects the first correction amount. A second correction amount commensurate with the longer predicted time is selected.

The fuel supply correction means corrects and controls the fuel supply amount based on a correction amount selected based on whether or not correction amount calculation is performed during a fuel supply period synchronized with engine rotation.

<Example> The present invention will be explained in detail below.

In FIG. 2 showing the system configuration of one embodiment, an internal combustion engine 1 includes an air cleaner 2. Air is taken in through the intake duct 3, the throttle chamber 4, and the intake manifold 5. The air cleaner 2 has an intake air (atmospheric) temperature TA (
An intake temperature sensor 6 for detecting "C) is provided.

The throttle chamber 4 is provided with a throttle valve 7 that operates in conjunction with an accelerator pedal (not shown) to control the intake air flow i1Q. The throttle valve 7 is provided with a potentiometer that detects its opening degree TVO, and an idle switch 8 that is turned on when it is in its fully closed position (idle position).
A throttle sensor 8 including A is attached.

The intake manifold 5 downstream of the throttle valve 7 is provided with an intake pressure sensor 9 that detects the intake pressure PB.
An electromagnetic fuel injection valve 10 is provided for each cylinder (four cylinders #4 to #4 in this embodiment).

The electromagnetic fuel injection valves 10 are intermittently driven to open by drive pulse signals outputted from a control unit 11 built in a microcomputer (to be described later) in synchronization with the intake stroke of each cylinder, and inject fuel (not shown) into the valves. Fuel that is pressure-fed from the pump and controlled to a predetermined pressure by a pressure regulator is injected and supplied into the intake manifold 5. That is, the amount of fuel supplied by the fuel injection valve 10 is controlled by the valve opening driving time of the fuel injection valve 10.

Further, a water temperature sensor 12 is provided to detect the cooling water temperature Tw in the cooling jacket of the engine l, and the exhaust passage 1
An oxygen sensor 14 is provided within the engine 3 for detecting the air-fuel ratio of the engine intake air-fuel mixture by detecting the oxygen concentration in the exhaust gas.

The control unit 11 receives a crank reference angle signal REF which is output from the crank angle sensor 15 at every predetermined crank angle position (every 180 degrees in the case of four cylinders, and in this embodiment, the intake BTDC of each cylinder is 90 degrees). Normal fuel supply is controlled using the reference angle signal REF as a reference position, and the engine rotational speed N is detected by measuring the cycle of the reference angle signal REF.

Incidentally, among the reference angle signals REF, the intake B of the #l cylinder
The one corresponding to TDC90' can be distinguished from the others, so that the reference angle signal REF can be made to correspond to each cylinder.

In addition, a transmission attached to the engine l is provided with a vehicle speed sensor 16 for detecting vehicle speed and a neutral sensor 17 for detecting a neutral position, and these signals are input to the control unit 11. Further, a voltage signal from a battery 20, which is a power source for opening the fuel injection valve 10, is input to the control unit 11 via an ignition switch 21.

Additionally, an auxiliary air passage 18 bypassing the throttle valve 7
is provided with an electromagnetic idle control valve 19 that controls the idle rotation speed via the amount of auxiliary air.

The control unit 11 controls the fuel injection amount Ti corresponding to steady operation based on the detection signals from the various sensors described above.
(during the pulse of the injection pulse signal), QB sets a correction amount corresponding to transient operation, and reduces the fuel supply amount Ti during deceleration operation to set the final normal fuel injection amount MTi. .

Then, the fuel injection valve 10 is individually controlled to open based on the set fuel injection amount MTi in accordance with the intake stroke of each cylinder to control normal fuel supply, while the normal fuel supply is controlled during acceleration. Following the spinning, a corrected amount of fuel is additionally supplied in accordance with the acceleration.

Historically, the control unit 11 performs feedback control of the idle rotation speed to a target idle rotation speed by controlling the opening degree of the idle control valve 19 during idle operation detected based on the idle switch 8A and the neutral sensor 17. do.

Next, skit 1. The various calculation processes for fuel supply control performed by the fuel unit 11 will be explained according to the routines shown in the flowcharts of FIGS. 3 to 7, respectively.

In this embodiment, the functions of the change amount calculation means, the first predicted time calculation means, the second predicted time calculation means, the correction amount calculation means, the correction amount calculation execution determination means, and the correction amount selection means 5 as the fuel supply correction means are 2. The software is provided as shown in the flowcharts of FIGS. 3 to 7. Further, in this embodiment, the operating state detection means includes a throttle valve 70 that variably controls the opening area of the throttle chamber 4.
A throttle sensor 8 that detects an opening of 1゛VO, and a crank angle sensor 15 that outputs a detection signal synchronized with engine rotation.
corresponds to

The routine shown in the flowchart of FIG. 3 calculates the fuel injection fi Ti corresponding to the steady state operation, and also calculates the correction amount DLTTP l to Dt, 'r' according to the transient operation state.
This is a routine that calculates rps, and is executed every 1.0ms.

First, in step 1, the throttle valve 70 opening degree TVO detected by the throttle sensor 8 is inputted, and in the next step 2, the intake pressure PB detected by the intake pressure sensor 9 and The engine rotation speed N calculated based on the detection signal from the crank angle sensor 15 is input.

In step 3, the throttle sensor 4 (of the engine intake system), which is variably controlled by the throttle valve 7, is opened [to the blood volume m+
12 in the slot 1. entered in step l. [.

This is determined by searching from a map set at -t based on the valve opening degree TVO.

In step 4, based on the value obtained by dividing the opening area A obtained in step 3 by the engine rotational speed N, the basic volumetric efficiency QHφ (%) of the engine 1 corresponding to steady operation is searched from the map.

In step 5, the intake pressure 1') entered in step 2 is
The value obtained by multiplying B by the engine rotational speed N (intake air flow rate Q
Equivalent (i) Search and find the weighted weight 2 from a preset map based on the weighted weight. This weighted weight 2 is used in the weighted average calculation of the volumetric efficiency Q CY L in the next step 6. .

In step 6, the basic volumetric efficiency Q Hφ obtained by searching in step 4 and the volumetric efficiency QCYL calculated in step 6 at the previous execution of this routine (10 m3 ago) are searched and obtained in step 5. Using 2 as the weighted weight, the volumetric efficiency QCYL is updated and set by weighted averaging according to the following formula.

Q CY L ← QH φ (1 - to 2) + QCYL ¥ to 2 The above weighted average calculation calculates the volume that approximately follows the actual engine load change that is delayed by C with respect to the change in opening area A with almost no detection delay and engine rotational speed N. This is for setting the efficiency QCYL.

In step 7, the basic monthly injection amount αNTp (engine load N) is calculated based on the volumetric efficiency QCYL according to the following formula.

αNTp←KCONAXQCYLXKTAHere, KC
ONA is a constant based on the injection characteristics of the fuel injection valve 10, KT
A is an intake temperature correction coefficient (air density correction coefficient) that is set by searching from a map based on the intake air temperature TA detected by the intake air temperature sensor 6 in step 42 of the background job (BGJ) shown in FIG. It is.

The basic fuel injection amount αNTp calculated here is calculated in order to detect the amount of change in engine load during transient operation and set the transient correction lid, and corresponds to normal fuel supply. The basic fuel injection amount TP is calculated based on the intake pressure PB, as will be described later. Here, the basic fuel injection amount Tp, which is set based on the detected values of the intake pressure PB and the intake air flow rate Q, is also a parameter for determining the engine load change amount, but the intake pressure PB and the intake air flow rate Q are used to pick up intake pulsations. Therefore, the basic fuel injection amount αNTp is determined based on the opening area A and the engine rotational speed N, since the true engine load change amount cannot be determined with high accuracy.

In the next step 8, the basic fuel injection amount αNTp calculated in step 7 during the previous execution of this routine is calculated from the latest basic fuel injection amount αNTp calculated in step 7 this time.
By subtracting poto, the amount of change ΔαNTp (amount of change in engine load) in the basic fuel injection amount αNTp per routine execution cycle (IOms) is determined.

In step 9, the amount of change Δα in the next step 8 is
In order to use it for calculating NTp, the basic fuel injection amount αNTp calculated in step 7 this time is used as the previous value αN T P
Set to OLD.

In step 10, the cycle time TRE of the reference angle signal REF output from the crank angle sensor 15 every 180° is
TMloMs indicating the elapsed time from the recent reference angle signal REF from F (time required to rotate 180°)
By subtracting ON, the time TATIME (see FIG. 8) from the current execution of this routine to the next reference angle signal REF is predicted and set.

Then, in the next step 11, the time TATIME
The correction amount DLTTpl is calculated using the amount of change ΔαNTp according to the following formula.

DLTTpl← Δa N T p X (TATIME+TREF
X/180) (For example, the intake ATDC1 of each cylinder
00"~intake BDC) from the reference angle signal REF to X'
(If the target is intake BDC, it is assumed that the crank angle position is 90@).

Therefore, for example, in the example shown in FIG. 8, when this routine is executed (1011s J OBON
DLTTp 1 in ) corresponds to the amount of change in engine load from that point up to the target crank angle position in the intake stroke of the #1 cylinder, for which normal fuel injection has already been completed. That is, the time until the next reference angle signal REF is TATIME, and the time until the next reference angle signal REF is TATIME.
The time from EF to the target crank angle position is TREFXX
/180, the time from the current execution of this routine to the nearest target crank angle position is TATIM
E+TREFxX/180, and ΔcrNTp is 10
Since it is the amount of change per m5, by multiplying the value obtained by multiplying this by 1/10 by the above-mentioned time, the amount of change in the basic fuel injection amount αNTP up to the nearest target crank angle position can be predicted and calculated. .

In the next step 12, it is determined whether the current execution timing of this routine coincides with the normal fuel supply by determining whether the flag FINJON is 1oro. As described later, the flag FINJON is set to 1 in synchronization with the start of normal fuel injection, which is performed in synchronization with the rise of the reference angle signal REF, and is set to 0 when the normal fuel injection ends. It looks like this.

Here, it is determined that the flag FINJON is zero,
At the current execution timing of this routine, if normal fuel supply is not being performed to any cylinder, the routine proceeds to step 13. In step 13, the time TATIME calculated in step 10 is set to the time MTATIME that is not updated when the flag FINJON is 1, and a flag is set so that it is determined that this routine has been executed during non-normal injection. Add a zero cent to Fk. Therefore, when this routine is executed when normal fuel injection is not performed, TAT IME = MTATI
ME, and the flag Fk becomes zero.

On the other hand, if it is determined in step 12 that the flag FINJON is 1, and normal fuel supply is being performed in any cylinder at the current execution timing of this routine, the process advances to step 14. In step 14, the time M
The update setting of TA'FIME is not performed, but only the process of setting the flag Fk to 1 is performed.

Therefore, if the current execution timing of this routine coincides with normal fuel supply, this routine will change the time MTAT.
The IME is not updated, and the time MTATIME retains the value of MTΔ''FrME at the previous execution of this routine.

In the next step 15, the time MTAT is calculated according to the following formula:
IME-based correction 1] Calculate DLTTp2.

1) LTTp2← Δcr N T p X (MTATIME+TREF
(1+X/180)) ) is longer by one period numerator REF, but the time until the next reference angle signal REF is MT
Since ATIME is used, when the current time MTATIME is not updated, the correction amount corresponds to the time from the previous execution of this routine to the second target crank angle position, and when the current time MTATIME is updated. The correction amount corresponds to the time from the current point to the second target crank angle position (the target crank angle position of the cylinder where normal fuel supply starts next) (see FIG. 9).

In the next step 16, in the same way, the time MTATIMH
Based on (correction amount DL-f''Tp3 is calculated according to the following formula.

DLTTp3← ΔαN T p X (MTATIME+TREF(2
+χ/180)) X 1/10 This correction amount DLTTp
3 (second correction amount) is the above correction! In addition to DLTTp2, 1 of the reference angle signal REF (normal fuel supply control)
This is the correction amount corresponding to the prediction time (second prediction time) that is longer by the periodic molecule REF, and the current time MTAT IM
If E is not updated, the correction amount corresponds to the time from the previous execution of this routine to the third target crank angle position, and when the current time M T A T I ME is updated, The correction amount corresponds to the time until the target crank angle position (the target crank angle position of the second cylinder from which normal fuel supply starts from the current time).

In step 17, a flag F10ms is turned ON so that execution of this routine during the fuel supply cycle will be determined later.
, F10msON2 are set to 1, and a counter value cnt for counting the number of times the normal fuel supply ends during this routine execution cycle is reset to zero, as will be described later.

In step 18, a correction amount DLTTp4 based on the time TATIMH, which is updated each time this routine is executed, is calculated according to the following 2.

D L T T p 4← Δa N T p X (TATIME+TREF(1
+X/180)) X l/10 above correction 1DLTTρ
4 is the correction amount (first correction amount) corresponding to the time from the current moment to the second target crank angle position (first predicted time), and when the time MTATIME is updated in step 13 this time, DLTTp4=DLTTp2. (8th
(see figure).

Further, in step 19, the correction amount D based on the time T A T I M B which is similarly updated every time this routine is executed is calculated.
LTTp5 is calculated according to the following formula.

DI-TTp5← ΔαN T p X (TATIAE -!-TREF
(2±X/180)) X 1/10 above correction 1iD
LTTp5 is a reference angle signal RE with respect to DLTTP4.
Since only one period numerator REF of F is added, the correction amount (second correction amount) corresponds to the time from the current moment to the third target crank angle position (second predicted time), and this time, in step 13, the time MTAT When the IME is updated, DLTTp 5=DLTTp 3 (see FIG. 8).

In step 20, the amount of change ΔαN calculated in step 8 is
By comparing Tp with a predetermined value, it is determined whether the engine 1 is accelerated, steady, or decelerated.

The amount of change ΔαNTp is the latest basic fuel injection amount αN
'l' Since it is the amount obtained by subtracting the previous value from p, the engine l
When is being accelerated, the amount of change ΔαNTP is a positive value, and conversely, when is being decelerated, the amount of change Δα
NTp becomes a negative value. Further, when the engine 1 is in steady operation, the absolute value of the amount of change ΔαNTp should be approximately zero.

Therefore, when the amount of change ΔαNTp is larger than a predetermined positive value, acceleration operation of the engine 1 is determined, and the amount of change Δ
When αNTp is smaller than a predetermined negative value, the engine 1 determines deceleration operation, and if the absolute value of the amount of change ΔαNTp is less than or equal to a predetermined value, the engine 1 determines steady operation.

When it is determined in step 20 that the engine 1 is decelerating, the process proceeds to step 21 and a deceleration flag F_dee for determining decelerating operation is set to 1. Further, when it is determined that the engine 1 is in steady operation, the process proceeds to step 22, where the deceleration flag F dec is set to zero, and the acceleration flag F ace for determining acceleration operation is also set to zero.

Further, when the acceleration operation of the engine 1 is determined in step 20, the acceleration flag F acc is determined in step 23, and when the acceleration flag F acc is zero and this is the first acceleration determination, the acceleration flag F acc is determined in step 24. acc
After setting 1 to 1, the processing in steps 25 to 29 causes the normal fuel supply control to execute asynchronous interrupt injection (additional supply).

In step 25, cylinder discrimination (ffi CY L N
The cylinder to which interrupt injection is to be performed is determined based on O. The cylinder discrimination value CYLNO is set to 1 at the normal fuel injection start timing of the #1 cylinder, for example, and is updated to 3 when the injection start timing of the next injection cylinder, the #3 cylinder, comes. The number corresponds to the cylinder number to which interrupt injection is to be performed.

Then, for the fuel injector 10 of the cylinder discriminated based on the cylinder discrimination value CYLNO in step 25, as shown in FIG. Output the drive pulse signal in the pulse corresponding to 2+Ts (step 26~
29). For convenience, the basic fuel injection amount αNTp is twice the correction amount DL”rTPl because it corresponds to the required fuel amount per cylinder. Also, Ts
is a correction amount for correcting a change in the effective injection time of the fuel injection valve 10 due to a voltage change of the battery 20 which is the driving power source of the fuel injection valve 10.

In addition, when the first acceleration determination coincides with normal fuel supply, the interrupt injection in steps 26 to 29 is not executed, and the shortage due to acceleration is supplemented by the additional supply that is performed after the normal fuel supply ends. to be done.

In step 30, the intake pressure PB input in step 2 is
The volumetric efficiency basic correction coefficient KPB is searched from the map based on .

Then, in the next step 31, a basic fuel injection amount 'rp PB based on the intake pressure PB is calculated according to the following equation.

'rpPB← C0NDXKPBXPBXKTAXKFLAT where, C0ND is a constant based on the injection characteristics of the fuel injector 10, KPB is the volumetric efficiency basic correction coefficient obtained in step 30, and KTA is set in step 42 of the background job shown in FIG. The intake air temperature correction coefficient FLAT is also a small correction coefficient obtained by searching from a map based on the intake pressure PB and the engine rotational speed N in step 41 of the background job shown in FIG.

In step 32, various corrections are made to the basic fuel injection amount TpPB obtained in step 31 according to the following formula to set fuel injection (]Ti corresponding to steady operation.

Ti← 2 XTp PBXCOEFXI, 8MBD^+
Here, C0EF is various correction coefficients mainly set based on the cooling water temperature Tw detected by the water temperature sensor 12, and LAMBD^ is the target air-fuel ratio determined through the exhaust oxygen concentration detected by the oxygen sensor 14. The air-fuel ratio feedback correction coefficient Ts, which is set to approximate the air-fuel ratio, is a correction based on the voltage of the battery 20.

In the next step 33, the value of the count value cnt2 counted up during the execution cycle of this routine is set to Mc
nt2, and in the next step 34 the count value cn
Reset t2 to zero. The count value cnt2 is incremented by 1 for each reference angle signal REF according to the routine shown in the flowchart of FIG.
0ms) indicates the number of reference angle signals REF.

Next, the routine shown in the flowchart of FIG. 5 is a routine for controlling the normal fuel supply for each cylinder, and is executed every time the reference angle signal REF is output from the crank angle sensor 15.

First, in step 51, the value of the timer Tref, which is incremented by 1 each time the routine is executed in step 92 of the routine shown in the flowchart of FIG. 6, which is executed every 1 ms, is set to the cycle time TREF of the reference angle signal REF, In the next step 52, the timer Tref is reset to zero. That is, the timer Tref is reset to zero when the reference angle signal REF is output, and then the timer Tref is reset to zero when the reference angle signal REF is output.
When the reference angle signal REF is output again, the count up result up to that point is TRE.
This is set to F.

Further, in step 53, the timer TMIOMSON is reset to zero, so that the timer TMIOMSON measures the elapsed time from the latest reference angle signal REF. Timer TMIO reset to zero here
MSON is 1ms, similar to the timer Tref.
In step 91 of the routine shown in the flowchart of FIG. 6, which is executed every time the routine is executed, 1 is added every time the routine is executed.

In the next step 54, the reference angle signal REF output from the crank angle sensor 15 this time is set to the intake BT of the #1 cylinder.
It is determined whether the output corresponds to DC90° or not.

The current reference angle signal REF is the #1 cylinder intake BTDC.
If it is determined that the angle is 90°, the process proceeds to step 55, where the deceleration flag F dec is determined. Deceleration flag F dec
When is 1, it indicates that deceleration operation of engine 1 was determined when the routine shown in the flowchart of FIG. state or is in an accelerated driving state.

In this embodiment, when the engine 1 is in a deceleration operation state, the normal fuel injection amount Ti is corrected to be reduced in accordance with the deceleration operation, whereas when the engine 1 is in an acceleration operation state, the normal fuel injection amount Ti is
is injected as it is, and the fuel that is insufficient due to acceleration is additionally supplied after the end of this normal fuel injection. Therefore, when it is determined in step 55 that the deceleration flag F dec is zero, the fuel is injected in step 56. Then, the fuel injection amount Ti calculated most recently in step 32 is set to MTi as the final fuel injection amount.

On the other hand, when it is determined in step 55 that the deceleration flag F dec is 1, the engine 1 is in deceleration operation and the normal fuel injection amount Ti should be reduced in accordance with the deceleration operation, so step 57 Proceed to the following.

In step 57, it is determined whether the flag F10ms is ON.

This flag F10msON is set to 1 when the routine shown in the third figure is executed, and is reset to zero when the reference angle signal REF is output in step 89, which will be described later. Therefore, when the flag F10msON is 1, the previous reference During the period from the angle signal REF to the current reference angle signal, in other words, during the normal fuel supply start timing, the routine shown in the third figure has been executed at least once. Therefore, in this case, The time from when the routine shown in the third figure was most recently executed to the target crank angle position of the #1 cylinder is TAT I ME + TRE F (1'+
X/180) (see FIGS. 8 and 9), the process proceeds to step 58 and the time (first
Prediction time) TAT IME+TREF (1+X/18
2xDLTTp4 (first correction amount) corresponding to 0) is subtracted and set as the final fuel injection amount MTi.

Further, when it is determined in step 57 that the flag F10msON is zero, the routine shown in the third figure is not executed during the recent normal fuel supply start timing, and when the previous reference angle signal REF was output (the previous normal injection start timing). This means that the routine shown in the third figure has been executed previously. In this case, the time from when the third illustrated routine was recently executed to the target crank angle position of the #1 cylinder is determined by the flag F.
For when 10msON is 1, the reference angle signal RE
F is extended by one period molecule REF (normal fuel supply period time) and TAT IME+TREF (2+X/180
), the process proceeds to step 59 and calculates the time (second predicted time) TATIME+TREF from the fuel injection amount Ti.
2XDLTTρ5 (second
(correction amount) is subtracted and set as the final fuel injection amount MTi.

As mentioned above, the required correction amount differs depending on whether or not the routine shown in the third diagram, that is, the routine for calculating the normal fuel supply amount and correction amount, has been executed between the recent reference angle signals REF. In the routine shown in Figure 3, the correction amount DLTTp5 (second correction amount) is calculated based on the prediction time [second prediction time] which is longer by one period numerator REF of the reference angle signal REF, and the correction 1DLTTp5 is calculated in advance. If so, the engine 1 will be operated at high speed, and the calculation frequency per certain rank angle will be reduced, and the reference angle signal R will be
Even when the correction amount is not calculated and updated between EFs, the normal fuel injection amount Ti can be corrected based on the correction amount that meets the request. However, such control can save memory capacity compared to setting the correction amount for each cylinder, and
Since there is no need to shorten the execution cycle of the correction amount calculation routine shown in FIG. 3, there is no need to increase the CPU load.

In this way, when the engine 1 is in deceleration operation, the fuel injection amount Ti to be used for normal fuel supply this time is
When a reduction correction is applied to the fuel injection MTi according to the amount of change (amount of decrease) in the engine load predicted in the time from when is set to the target crank angle position of the #l cylinder, the next step 60 is performed. In #I cylinder fuel injection valve 10
A drive pulse signal with a pulse width equivalent to MTi is output to the engine to start normal fuel supply to the #1 cylinder.

Since normal fuel injection has started, the next step is step 61.
Now set the flag FINJON to 1 and set the flag FI
It is determined by NJON that normal fuel supply is in progress.

Also, this time we started normal fuel supply to the #1 cylinder, so we set the cylinder discrimination value CYLNO to 1 and
Cylinder discrimination 4K CY L NO
Make it possible to distinguish by

On the other hand, in step 54, the current reference angle signal REF is #
If it is determined that it does not correspond to the intake B TDC 90° of the first cylinder, the process proceeds to step 63 and it is determined whether the current reference angle signal REF corresponds to the #2 cylinder.

If it corresponds to the reference position of the #2 cylinder, the #1
Similarly to the case of cylinders, flag F is set during deceleration operation.
The final fuel injection amount MTi is set by selecting the reduction correction amount based on the fuel setting timing determined by 10msON and reducing the fuel injection amount Ti (steps 64 to 71). Since it is the same as the control for , the explanation will be omitted.

Similarly, if it is determined in step 63 that the reference angle signal REF is not for the #2 cylinder, then step 72
It is determined whether or not the reference angle signal REF corresponds to cylinder #3, and if it is the reference angle signal REF for cylinder #3, then #3
Normal fuel supply control to cylinders (steps 73 to 8)
0). Further, in step 72, the reference angle signal R of the #3 cylinder is
If it is determined that it is not EF, the reference angle signal is REF for the remaining #4 cylinder, so if it is determined No in step 72, normal fuel supply control including deceleration reduction correction is performed for the #4 cylinder (step 81-88).

In this way, if the fuel supply is controlled depending on which cylinder the current reference angle signal REF corresponds to, and normal fuel supply control is performed with timing matched to the intake stroke of each cylinder, then in step 89 the above-mentioned A count value cn that resets the flag F10msON to zero and counts the number of reference angle signals REF (normal fuel supply start).
Increase t2 by 1.

Next, the routine shown in the flowchart of FIG. 7 is executed when normal fuel supply ends (T i end) with the reference angle signal REF as the supply start timing controlled by the routine shown in the flowchart of FIG. It is something that

First, in step 101, an acceleration flag F acc which is set based on the amount of change ΔαNTp is determined according to the flowchart of FIG. And acceleration flag F acc
During acceleration operation when is 1, the process proceeds to step 102 and subsequent steps, and following the end of normal fuel supply, a correction amount corresponding to the increase in engine load during acceleration is additionally supplied.

In step 102, a flag PIOmsON2, which is set to l every time the flowchart in FIG. 3 is executed, is determined, and the routine (normal fuel supply amount) shown in the flowchart in FIG. and correction amount calculation routine) have been executed. In other words, in the same way as when a correction amount whose predicted time differs by TREE is selected based on the flag F10msON during deceleration, when accelerating, the normal fuel supply end Tie is selected.
The amount of correction is selected depending on whether or not the routine shown in the third figure is executed during the period of nd.

Here, if it is determined that the flag F 10 m s ON 2 is zero and the third illustrated routine has not been executed during the recent end of normal injection, the routine proceeds to step 103.

In step 103, the flag Fk is checked to determine whether or not the timing at which the third illustrated routine was executed before the end of the previous normal fuel supply was during normal fuel injection. Here, if it is determined that the flag Fk is zero (for example, if the flag Fk of cylinder #4 in FIG.
In step 104, DLTTp5 (second correction amount) is selected as the correction amount M to be additionally supplied following the normal fuel supply.

That is, in the above case, the routine shown in the third diagram is executed before the start of the normal injection that was performed before the normal fuel injection that ended this time, and for the cylinder that ended the injection this time, at that time, This means that normal injection is controlled based on the calculated fuel supply amount Ti. Therefore, the correction amount corresponding to the engine load increase change during acceleration is the change amount ΔαNT calculated at the same time when the fuel injection amount Ti is finally calculated.
p and the time from the calculation of the amount of change ΔαNTp to the target crank angle position of the current injection end cylinder, and the said time is TAT IME + TRE
Since F (2+X/180) (second predicted time),
As the correction amount, DLTTp5 (second correction amount) is selected.

On the other hand, when it is determined in step 103 that the flag Fk is 1, if the routine shown in the third figure was executed during the fuel injection performed before the fuel supply that ended this time (the routine shown in the ninth figure).
#4 cylinder or #2 cylinder), and in this case, the process advances to step 105 to determine the count value cnt.

The count value cnt is reset to zero when the routine shown in FIG.
It is something that is raised.

Therefore, in the example shown in FIG. 9, for example, the #4 cylinder Ti
At the end, the count value cnt is the Tiend of the #3 cylinder.
Since the count is 1, the process proceeds from step 105 to step 106 and DLTTp4 is calculated as the correction iM.
Select.

Also, at the time of #2 cylinder Tiend in Fig. 9, #
Since the Tiends of the 3rd cylinder and the #4 cylinder are counted, cnt=2, and steps 105 to 10
Proceed to step 7 and select DLTTp5 as the correction amount M.

That is, in this case as well, the fuel injection amount T of the cylinder where injection has ended is
The correction amount is selected based on the time from when i is set to the target crank angle position of the cylinder where the current injection has ended.

Further, if it is determined in step 102 that the flag F10msON2 is 1, and the fuel injection amount calculation routine has been executed between the end of the most recent normal injection, the routine proceeds to step 108.

In step 108, the flag Fk is determined.

Here, when the flag Fk is zero and the routine shown in the third figure executed between the end of the recent normal injection is executed in a state where normal fuel injection is not performed (for example, #3 in FIG. (in the case of a cylinder), the process proceeds to step 109 and selects DLTTp4 (first correction amount) as the correction iM.

On the other hand, if it is determined in step 108 that the flag Fk is 1, the fuel injection amount calculation routine (routine shown in the third figure) executed during the recent end of normal injection is applied to the normal fuel injection that ended this time. This is a case where the executions overlap.

At this time, as in the case of #3 cylinder in the example shown in FIG. 9, when the routine shown in the third diagram is executed during injection,
The time TAT I ME is updated, but #3
The fuel injection amount Ti used for normal fuel supply to the cylinder is not the one set in the routine executed during normal injection, but the fuel injection amount T set in the previous routine.
i is used.

Therefore, when additionally supplying fuel in accordance with the acceleration state, the fuel injection for the normal fuel supply of the #3 cylinder is not used as a reference, rather than the routine shown in the third figure executed during the injection of the #3 cylinder. The engine load change amount should be predicted based on the time from when the amount Ti is set to the target crank angle position of the #3 cylinder. Therefore, in this embodiment, as described above, when the execution time of the routine shown in the third figure coincides with normal fuel injection, the update is not performed, and when the routine is executed during non-injection, the time T
A time MTAT IME is provided in which AT IME is set, and in the example shown in FIG. The time until the angle signal REF is TATIME, and MTATIME is also set to the same time. However, when the next routine execution coincides with the #3 cylinder injection, TATIME will be updated, but MTATIME will contain the previous time. The value of TATIME is set without being updated.

When it is determined in step 108 that the flag Fk is 1 and the process proceeds to step 110, a determination is made of Mcnt2 in which the number of reference angle signals REF counted during the execution of the routine shown in the third figure is stored.

Here, when Mcnt2 is 1, for example, in the case of cylinder #3 in Fig. 1O, the process proceeds to step IN and the correction amount M
DLTTp2 is selected as DLTTp2. This DLTTp2 is
This is calculated when the third illustrated routine is executed during normal fuel injection, but TAT rME7'4;! is used to calculate the time to the target crank angle position. Naku MTA
In the example shown in Fig. 10, T I ME is set based on the time from when the normal fuel injection amount Ti of the #3 cylinder is calculated to the target crank angle position of the #3 cylinder. Ru.

However, although the previous value is used for the time to the target crank angle position, the latest data obtained during normal injection is used for the amount of change ΔαNTp, and the correction amount is adjusted according to the more recent acceleration state. As set.

Further, when it is determined in step 110 that Mcnt2 exceeds 1, for example, in the case of cylinder #3 in FIG.
t2=2), at least the reference angle signal RE
Since the prediction period will be extended by one cycle of F, the process proceeds to step 112 and DLTTp3 is selected as the correction iM.

In this way, in this embodiment, the target time M and T A T I M E which are not updated during normal injection are provided, so that even if the routine shown in the third figure is executed during normal injection, the target time M and T A T I M E are not updated during normal injection. Since the correction amount is set based on the time since the final setting of the fuel injection amount Ti before the start of normal injection, the correction amount commensurate with the predicted engine load change is added following the end of normal fuel injection. In supply control, fuel correction controllability during acceleration is improved.

In the case where the third illustrated routine has been executed during the recent normal fuel injection, and after selecting the correction amount M in step 111 or step 112, the process proceeds to step 113, and the target updated by the execution during the normal injection is Time TAT I M
Set E to MTAT I ME.

As shown in Fig. 10, this occurs during normal fuel injection.
If the illustrated routine (10ms JOB) is executed and the execution timing of the next third illustrated routine coincides with the next normal fuel injection, the normal injection of the #4 cylinder will be the same as the normal injection of the previous #3 cylinder. This is because the additional correction amount for the #4 cylinder should be set based on the target time TAT I ME calculated during the normal injection for the #3 cylinder.

If the process in step 113 is not performed, the target time MTATMME will not be updated until the routine shown in the third figure is executed in a state where normal injection is not performed, and the target time MTATMME will not be updated until the routine shown in the third figure is executed in a state where normal injection is not performed. state is 2
If this continues more than once, the desired engine load change cannot be predicted.

As described above, during acceleration, the correction amount M is adjusted according to the increase in the engine load from the time when the fuel injection amount T i is set before the start of normal injection to the target crank angle position of that cylinder.
When the cylinder discrimination value CY is selected, in the next step 114, the cylinder discrimination value CY is selected.
Determine LNO. Then, the cylinder in which normal injection has ended is determined based on the value of the cylinder determination value cYLNO,
The drive pulse signal in the pulse corresponding to twice the correction amount M is outputted to the fuel injection valve 10 of the corresponding cylinder (steps 115 to 118), and is applied to the intake stroke of each cylinder in synchronization with the reference angle signal REF. Additional fuel is supplied during acceleration following the completion of normal fuel injection, which is performed at the same time.

Note that doubling the correction IM is similar to doubling the basic fuel injection amount 'r''pPB when setting the fuel injection amount Ti; the required amount when supplying fuel to each cylinder is This is because in this embodiment, the amount is set to be twice the calculated basic fuel injection amount αNTI).

After determining the cylinder where normal fuel injection ends and supplying an additional amount commensurate with the acceleration state, in the next step 119, since normal injection has ended, the flag FINJON, which is set to 1 at the start of normal injection, is reset to zero. At the same time, F is set to 1 when the third illustrated routine is executed.
10msON2 is reset to zero so that F10msON2 remains zero until the routine shown in the third figure is executed. Furthermore, a count value c for counting the number of times the normal injection ends during the execution cycle period of the routine shown in the third diagram.
Increase nt by l.

Further, when it is determined in step 101 that the acceleration flag F acc is zero and the engine 1 is in a steady operation or deceleration operation state, the engine jumps from step 101 to step 119 and does not perform additional supply following the end of normal injection.

In this embodiment, although the normal fuel injection amount is performed based on the intake pressure PB, an air flow meter is provided instead of the intake pressure sensor 9. The normal fuel injection amount may be set based on Q. Further, in the present embodiment, normal fuel supply is controlled with the injection start timing constant, but the injection start timing may be variably controlled so that the injection end timing is a constant timing.

<Effects of the Invention> As explained above, according to the present invention, the engine load change amount is calculated in synchronization with the calculation of the normal fuel supply amount, and the time to a predetermined crank angle position is predicted, and this time Set the time by adding the fuel supply cycle time to . Then, the predicted amount of change in engine load at each of the above times is calculated, two correction amounts commensurate with each predicted change amount are calculated, and depending on whether or not the correction amount has been calculated and updated during the fuel supply control cycle. The fuel supply amount is corrected by selecting one of the two correction amounts.

As a result, even at high engine speeds when the frequency of correction amount calculation for a constant crank angle is low, prediction accuracy of the engine load change amount is ensured, and air-fuel ratio controllability during transient operation in the high speed range is improved. However, there is an advantage in that obtaining such an effect does not involve an increase in memory capacity or an increase in the calculation load on the CPU.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of the present invention, Fig. 2 is a system schematic diagram showing an embodiment of the invention, Figs. 10 to 10 are time charts for explaining the control characteristics in the above embodiment, respectively, and FIG. 11 is a time chart for explaining the problems of conventional control. 1... Engine 4... Throttle chamber 7...
Throttle valve 9...Intake pressure sensor Trol unit 8...Throttle sensor 10...Fuel injection valve 11...C15...Crank angle sensor Patent applicant Japan Electronics Co., Ltd. Agent Patent attorney Sasashima Fujio

Claims (1)

[Scope of Claims] A fuel supply control device for an internal combustion engine configured to intermittently supply fuel at a timing synchronized with engine rotation, comprising: an operating state detection means for detecting a state quantity related to engine load; a change amount calculation means that is performed in synchronization with the calculation of the normal fuel supply amount performed at predetermined time intervals and calculates the amount of change in the engine load based on the detected state quantity; and the amount of change in the engine load is calculated. a first predicted time calculating means for predicting and calculating the time required from when the crank angle is reached to a predetermined crank angle position, and setting the required time as a first predicted time for calculating the predicted amount of change in engine load; a second prediction time setting means for setting a second prediction time by adding a cycle time of fuel supply to the first prediction time; Calculating a first correction amount of the fuel supply amount commensurate with the predicted amount of change in the engine load, and based on the amount of change in the engine load and the second predicted time, the predicted amount of change in the engine load at the second predicted time. correction amount calculation means for calculating a second correction amount of the fuel supply amount commensurate with the engine rotation; correction amount calculation execution determining means for determining whether or not the correction amount calculation means is executed during a fuel supply period synchronized with engine rotation; selecting the first correction amount when the correction amount calculation execution determining means determines that the correction amount calculation means is to be executed;
A correction amount selection means for selecting a second correction amount when it is determined that the correction amount is not to be executed; and a fuel supply correction means for correcting and controlling the fuel supply amount based on the selected correction amount. Fuel supply control device for internal combustion engines.