JP3331689B2 - Fuel injection control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electromagnetic spill fuel injection control device for supplying fuel to a diesel engine.

[0002]

2. Description of the Related Art Conventionally, as a technique in such a field, there is a fuel injection control device described in JP-A-61-223248 and JP-A-4-86352. In these fuel injection control devices, the high-pressure fuel pressurized by the plunger is caused to overflow by an electromagnetic valve at a timing of an appropriate desired crank angle (° CA: crank angle), thereby terminating the fuel injection, and injecting the fuel. The electronic control device is used to operate the solenoid valve by using a microcomputer. In the above-described solenoid valve spill system, it is necessary to precisely control the fuel end timing, that is, the valve opening timing of the solenoid valve, so as to be at a predetermined crank angle position corresponding to the fuel injection amount. However, this is because the fuel injection amount tends to fluctuate greatly.

In the devices disclosed in both the publications, the injection amount is obtained from the rotation angle (° CA) of the engine as described above, and the injection amount is adjusted by opening or closing the solenoid valve at that timing. The timing is classified into two types: a rotation angle pulse number (integer) counted from a predetermined reference rotation position and a set time between rotation angles. The former is obtained as the injection angle corresponding to the rotation angle, and the latter is obtained by angle-time conversion from the rotation angle time interval of the surplus angle.

[0004]

However, the conventional fuel injection control device described in Japanese Patent Application Laid-Open No. 4-86352 has the following problems. FIG. 8 is a time chart for explaining the conventional solenoid valve opening / closing timing. As shown in the figure, the opening angle of the solenoid valve at the end of the injection is between 4 and 5 of the rotation angle pulse. At this time, the injection angle No. Is 4
And the valve opening time T, which is the set time from the injection angle, is the surplus angle between the injection angle pulses, and is determined by the time T45 between the injection angles 4 and 5. Precisely, the valve opening set time T cannot be calculated unless T45 after the NE interruption of the rotation angle 5 is calculated. However, at that time, the valve opening time T has already passed and control is impossible. Therefore, in this prior art, 180 ° C
The time between the same rotation angle before A, that is, the cylinder is different, but the time of the same rotation angle of one of the cylinders such as compression and explosion, in this example, at the time interval of the rotation angle pulses 4 and 5 180 ° CA before this closing, The valve time T was determined from the surplus angle.

[0005] For this reason, the fuel injection control device of the prior art, when the engine rotational speed is constant, the current T4
It is possible to judge that T45 before 180 ° CA is almost the same as T45 before 180 ° CA. However, T45 before 180 ° CA sometimes greatly differs when there is a change in engine rotation, during acceleration, and during deceleration. FIG. 9 is a diagram showing the relationship between the valve opening command angle during acceleration or deceleration and the injection amount. As shown in the figure, the injection amount with respect to the valve opening command angle is represented by a straight line passing through the rotation angle pulse point at the time of stable rotation or at the time of ideal injection amount. On the other hand, when the rotation fluctuates during acceleration or deceleration or the like, an error occurs because the surplus angle time conversion is performed at a rotation angle time interval of 180 ° CA before. In particular, when the surplus angle is large and immediately before the rotation angle, the maximum error occurs. Therefore, if the valve opening command angle is the same as the rotation angle, that is, some rotation fluctuation occurs in a state where there is no surplus angle, and the valve opening angle fluctuates before and after the rotation angle pulse, the injection amount immediately before the rotation angle pulse Tend to become unstable during acceleration and become less stable during deceleration. Furthermore, in the diesel engine, there is a further problem that the governor pattern causes a large rotation fluctuation because the injection amount increases when the rotation speed decreases, and decreases when the rotation speed increases. For this reason, the valve opening command angle is devised so that the valve opening command angle does not fluctuate around the rotation angle pulse at the time of idling or the like. There is a problem that the state of going around the corner is inevitable.

Next, another prior art is disclosed in
In the fuel injection control device described in the fuel injection control device described in Japanese Patent Application Laid-Open No. 1-223248, the angle time conversion of the surplus angle is obtained at the immediately preceding pulse interval, but there are the following problems. Since the angle time conversion needs to be performed at the immediately preceding pulse interval, the injection angle No. Thus, the processing is performed within an NE interrupt as described later. Performing the conversion process before the necessity of setting the conversion time of the surplus angle in the interrupt prior to the necessity of setting the time and the time after the conversion are too small from the previous injection angle Attempting to set the time is impossible and impractical. Also, 1
Although it is considered that the step at the time of switching the injection angle of the present invention is slightly better from the immediately preceding pulse interval instead of the data before 80 ° CA, the 180 ° CA of the operation such as compression and explosion is considered.
In each cycle, the angle is different at the immediately preceding pulse interval, and there is a problem that an error occurs even during stable rotation.

That is, in the control system in which the angle of the injection quantity is calculated, the reference angle pulse number and the surplus angle are angle-time converted from the previous reference pulse time interval, and the time is set as the injection end signal when the reference angle pulse is input, the remainder angle time conversion is performed. The reference angle pulse time interval to be performed is the previous time interval within the same surplus angle.
Big . At this time, the deviation from the ideal injection amount is larger as the surplus angle (set time) is larger. It is necessary to reduce the deviation and the step at the time of switching the time set reference angle.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a fuel injection control device capable of reducing a step of an injection amount caused when an injection angle is changed due to a change in rotation and making a smooth connection. I do.

[0009]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention aims to control the fuel injection amount by overflowing high-pressure fuel by opening a solenoid valve at a desired timing. The spill angle corresponding to this injection amount is divided by the rotation angle output for each crank angle, the number of rotation angle pulses k, which is a quotient of integers, and the remaining angle are obtained. In the fuel injection control device for forming a signal, the first memory stores a time T ′ converted from a previous remainder angle into a time by linear interpolation. The second memory stores the time interval Tk−1 of the rotation angle pulse (K) immediately before the current injection.
Remember k. The third memory stores the time interval T'k-1, k of the rotation angle pulse (K) immediately before the previous injection. The fourth memory stores a correction coefficient for correcting the time T using a difference between the current time interval Tk-1, k and the previous time interval T'k-1, k as a parameter. The fifth memory stores a correction processing time T0 which is a maximum time required to correct the time T '. When the time T 'is equal to or longer than the correction processing time T0, a difference between the present time interval Tk-1, k and the previous time interval T'k-1, k is obtained, and a correction coefficient is further obtained from the difference. The correction coefficient is corrected by multiplying the correction coefficient by the time T ′ to derive the current time T to form an injection end time from the time T ′. On the other hand, when the time T ′ is less than the correction processing time T0, the correction is not performed from the time T ′. The injection end time is formed. The sixth processing which stores the processing time Ta required for the processing when no further correction is made
In the case of time T ′ <Ta, the rotation angle pulse (k) immediately before the current injection is converted to the rotation angle pulse (k−
1) and T '= T' + Tk-1, k.

[0010]

According to the fuel injection control device of the present invention, a difference between the present time interval Tk-1, k and the previous time interval T'k-1, k is obtained, and a correction coefficient is further obtained from the difference. Then, the correction coefficient is multiplied by the time T 'and corrected to derive the current time T to perform the correction processing. However, since the correction processing time T0 is required, it is first determined whether the time T' is longer than the correction processing time T0. In the case of an affirmative determination, the above-described correction is performed on the time T ′, and a time T that is close to true in consideration of acceleration or deceleration can be predicted. On the other hand, if the time T 'is less than the correction processing time T0, the effect of acceleration or deceleration is small and there is no need for correction, so the uncorrected time T' is used directly in the calculation of the injection end time. Thus, as in the conventional case where there is acceleration or deceleration, a step in the injection amount caused when the injection angle is changed due to a change in rotation can be reduced, and the connection can be made smoothly. If the time T ′ <Ta, the rotation angle pulse (k) immediately before the current injection is changed to the rotation angle pulse (k−1) and T ′
By setting = T '+ Tk-1, k, this correction processing can be performed even with a computer having a somewhat slow processing speed.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. An example of the solenoid valve spill system to which the embodiment of the present invention is applied will be specifically described with reference to the drawings. FIG. 1 is an overall configuration diagram of a solenoid valve spill system based on a Bosch distribution type fuel injection pump according to an embodiment of the present invention. In this figure, a drive shaft 1 driven by an engine (not shown) rotates a vane type feed pump 2, which introduces fuel from a suction port 3 and pressurizes the fuel. After adjusting the pressure to a predetermined pressure, the pressure is supplied to the fuel chamber 6 formed in the pump housing 5. The drive shaft 1 drives a pressure feed plunger 8 via a coupling 7. The coupling 7 rotates the pressure-feeding plunger 8 integrally in the rotation direction,
The reciprocating motion of the pressure-feeding plunger is freely allowed in the axial direction. A face cam 9 is provided integrally with the pressure feed plunger 8. The face cam 9 is pressed by a spring 10 and pressed by a cam roller 11. The cam roller 11 and the face cam 9 have a well-known configuration that converts the rotation of the drive shaft 1 into reciprocation of the pressure feed plunger 8, and the plunger 8 makes one rotation as the cam ridge of the face cam 9 rides on the cam roller 11 by sliding contact. The cylinder is reciprocated a number of times according to the number of cylinders. The pressure feed plunger 8 is fitted to a head 12 fixed to the housing 5 to form a pump chamber 13. The suction plunger 8 is formed with a suction groove 14. When one of the suction grooves communicates with the suction port 15 during the suction stroke of the pressure feeding plunger 8, fuel is introduced from the fuel chamber 6 to the pump chamber 13. When the fuel in the pump chamber 13 is compressed during the compression stroke of the pressure-feed plunger 8, a value amount is sent from the distribution port 16 to the fuel injection valve (not shown) of each cylinder through the pressure-feed valve 17, and is injected into the combustion chamber of the engine.

A fuel metering mechanism 20 is connected to the pump chamber 13. When the current of the coil 22 of the electromagnetic valve 21 is stopped , the needle valve 23 is lifted, and the fuel in the high-pressure pump chamber 13 is returned to the fuel chamber 6 through the overflow paths 24 and 25. It is configured in. Therefore, when stopping the work <br/> movement of the solenoid valve 21 of the fuel injection ends during the compression stroke of the pumping plunger 8. Here, the end of energization of the solenoid valve 21 is performed by an electronic control device 26 such as a microcomputer. The electronic control unit 26 receives a signal of an engine operating state detected by various sensors of the engine, for example, a temperature sensor 30, an accelerator pedal sensor 40, and the like, and a signal from a rotation angle sensor 50, and receives a signal from the rotation angle sensor 50. To control the energization of

FIG. 2 shows a rotation angle sensor 5 having a section taken along the line aa in FIG.
FIG. The rotation angle sensor 50 includes a disk 52 having a plurality of protrusions 51 integrally attached to the injection pump drive shaft 1 and a proximity detector 53 such as a known electromagnetic pickup for detecting passage of the protrusions 51. Disk 5
The two projections 51 are arranged at intervals of 90 ° / 16 = 5.625 °, and each of the discs 52 has a missing portion 54 corresponding to two projections every 90 °. At 90 ° rotation of the injection pump drive shaft 1, that is, at every 180 ° crank angle of the engine, the missing portion 54 is detected and used as a reference angle signal, and at every 11.25 ° crank angle of the engine, the projection 51 is detected and the rotation angle signal is detected. And

[0014] will be described with reference to the drawings a control method according to the present invention based on the above configuration. FIG. 3 is a timing chart showing the concept of the auxiliary operation of the injection amount adjustment by the electronic control unit. FIG. 7A shows the lift of the plunger of the injection pump, FIG. 7B shows the energization pulse signal to the spill adjustment solenoid valve 21, and FIG. 8C shows the signal from the rotation angle sensor 50. Computer (2
6) determines a fuel amount q to be injected based on load information from the temperature sensor 30, the accelerator sensor 40, a pressure sensor (not shown), and the like, and determines a spill angle θ corresponding to the injection amount q, that is, a reference angle θ °. After the lapse of the crank angle, the power supply to the spill solenoid valve 21 is stopped to terminate the injection. It is naturally necessary for the solenoid valve 21 to be closed again during the suction stroke of the plunger in preparation for the next injection. However, the closing may be performed during the suction stroke, and may be performed as compared with the valve opening side. The requirements for the timing accuracy need not be very small.

On the other hand, the spill angle θ ° CA, which is the spill solenoid valve opening timing, is an important parameter directly related to the injection amount, and needs to be controlled with extremely high precision. The accuracy of the spill angle θ ° CA can be easily secured if a rotation angle signal having infinitely fine resolution can be input.
At the current technical level, it is necessary to use a limited number of rotation angle signals and a time counter built in the control microcomputer in combination. In this embodiment, since the output interval angle θ ₀ of the rotation angle signal is 11.25 ° CA, a small angle can be obtained by interpolation. First, the spill angle θ determined based on the required injection amount q is divided by θ ₀ to obtain a quotient k (integer) and a remainder angle θh. Therefore, θ ₀ × k
While ° portion corresponding to the CA can be determined correctly angle signal, the angle of the minimum resolution theta ₀ too smaller than the angle signal theta
Since h can no longer be treated as an angle, it is converted into a time corresponding to the surplus angle θh based on the engine speed at that time. As described above, after k rotation angle signals have elapsed from the reference angle, and after the remaining angle θh has elapsed,
Command the spill solenoid valve to open.

Hereinafter, the point of view of the present invention will be described.
FIG. 4 is a view showing a rotation angle pulse signal and an opening signal of a spill adjustment solenoid valve according to movements during stable rotation, acceleration, and deceleration, in the viewpoint of the present invention. As shown in the figure, when the injection amount is between the reference angle and the rotation angle signals 4 and 5, the injection angle k = 4 and the remainder angle θh is equal to the rotation angle signal 4 before 180 ° CA.
The angle time is converted by linear interpolation from the time interval T45 of (5) and (5) to obtain T as the valve opening time from the rotation angle signal 4. The valve opening time T does not significantly change from the substantially same time interval between T45 before 180 ° CA and the current T45 before performing angle time conversion during stable rotation.

FIG. 5 is a view showing the injection amount with respect to the opening command angle of the spill adjustment solenoid valve in relation to the viewpoint of the present invention. In this figure, when the valve opening command angle is changed slowly enough to be considered as a stable rotation, the injection amount can be represented by a straight line connecting the injection points with respect to the rotation angle pulse signal. This is the injection amount for the ideal valve opening command angle. On the other hand, since the angle time is converted at the time interval of 180 ° CA before, the rotation angle signal point is accurate, but immediately before that, at the time of acceleration, the time interval before 180 ° CA is longer than this time. Since the injection amount is larger and the deceleration is shorter than this time from the time interval before 180 ° CA, the injection amount is smaller. In other words, a sawtooth waveform that passes through the injection amount corresponding to the rotation angle pulse number and immediately before the injection amount relative to the ideal valve opening command angle has the maximum deviation (FIG. 9).
Reference). This shift varies greatly depending on the amount of acceleration and deceleration, and increases as the surplus angle θh (valve opening time T) increases. Assuming that the time required for correcting this deviation is, for example, T0 = about 150 μs, the deviation at the valve opening time T ′ <T0 is immediately after the rotation angle signal and is small. This is because when the rotation angle signal interval and T0 = about 150 μs are calculated in proportion to the maximum deviation immediately before the rotation angle signal, the engine rotation = 1000 r
pm, about one-twelfth of the maximum deviation, about 3000 rpm at about one-fourth. An object of the present invention is to provide a rotation angle N for setting a time T obtained by converting a remainder angle during acceleration / deceleration into an angle-time conversion.
o. This is to reduce the step at the time of switching. This T
Is smaller than the time T0 required for correction within the NE interrupt, the deviation is also small, and therefore, for T 'equal to or longer than T0, the latest acceleration / deceleration data is used as shown in FIG. When the valve opening time is between the rotation angle pulses 4 and 5 and the time T is set by the rotation angle pulse 4, the value of 3 immediately before the time T is set.
And T4, which is 180 ° CA before that
And the step at the time of switching the rotation angle is predicted, and
The purpose is to correct the angle time conversion time T ′ of the remainder angle θh calculated at the time interval before ° CA to reduce the step at the time of switching the rotation angle and to connect smoothly. That is, from the difference between the current T34 and T34 (−180 °) before 180 ° CA, the next rotation angle No. 5 is predicted, and a correction coefficient is set so that T becomes equivalent to the current T45 remainder angle time conversion. Hereinafter, specific processing in the electronic control unit 26 will be described.

FIG. 6 is a flowchart for explaining the energization opening control of the solenoid valve 21 by the electronic control unit 26 of FIG. First, the main routine shown in the figure executes the calculation of the injection amount and the injection angle / remainder angle time conversion processing. In the main routine, in step 101, the accelerator pedal sensor 40 reads the accelerator depression amount. The accelerator pedal sensor 40 is based on a general potentiometer (not shown), and a sensor whose voltage changes according to the amount of depression of the accelerator is converted by an analog / digital converter (ADC) at a constant period (for example, every 8 ms). From this depression amount, what percentage the accelerator opening is is calculated.

In step 102, the engine speed (NE) is calculated from the 180 ° CA time interval obtained by the NE interrupt.
Is calculated. In step 103, a basic injection amount is calculated based on an injection amount pattern that is unique to the diesel engine and is determined by the engine speed and the accelerator opening, that is, a governor pattern.

In step 104, the final injection amount is corrected by adding corrections such as water temperature, intake air temperature, and supercharging pressure (not shown). In step 105, the final injection amount is calculated based on the injection time k (integer) as to which injection angle the injection angle corresponds to and the time T obtained by converting the remainder angle θh into angle time. The reason why the injection angle and the surplus angle time are converted in the main routine will be described later. Next, processing in the NE interrupt will be described.

In step 201, it is determined whether or not the injection angle k calculated in the main routine matches the present rotation angle. If not, the process proceeds to step 210 to perform another process. . In step 202, when the values coincide with each other in the above step, it is determined whether or not the time T 'converted from the remainder angle is longer than the correction processing time T0 added this time. When T ′ ≦ T0, the valve opening time of the spill metering valve 21 has passed within the correction processing time T0, so no correction is performed, and the routine proceeds to step 208, where the valve opening time T ′ from the current rotation angle is set. Is done. Electronic control unit 26
Is stored in the first memory and T0 is stored in the fifth memory.

In step 203, when T '> T0, the immediately preceding rotation angle, the time interval between the previous rotation angle and its 180 °
The difference between the same rotation angle time intervals before CA is taken. In this embodiment, when the injection angle is 4, the time interval T3 between the current rotation angles 3 and 4
4 (this time) and T34 (previous time) before 180 ° CA. In step 204, depending on whether the difference is positive or negative,
Look at acceleration or deceleration. For example, this time T34 <previous T34
If so, it can be determined that the vehicle is accelerating, and vice versa.

In steps 205 and 206, a correction coefficient is obtained from the direction of acceleration / deceleration and the amount of acceleration / deceleration of the difference obtained in step 203. This correction coefficient will be described below. FIG. 7 is a diagram for explaining a correction coefficient at time T ′ obtained by converting the remainder angle θh into angle time. As shown in the figure, the correction coefficient is obtained by a calculation formula or map interpolation according to the acceleration / deceleration amount. This correction coefficient corresponds to the injection angle No. In addition to the above, it is conceivable that the injection amount and the engine speed may be further provided as parameters if necessary. The correction coefficient is stored in the fourth memory of the electronic control unit 26.

In step 207, next, in step T20, the remainder angle time is converted to T 'by T45 before 180 ° CA.
The valve opening time T after the correction is multiplied by the correction coefficient obtained in steps 5 and 206. In step 208, T ′ is set to T if the determination in step 202 is negative. In step 209, T is set as the final valve opening time, this control ends, and the flow proceeds to the next control after step 210. The time T0 required for the above-described correction processing is the maximum required time of the processing from step 202 to step 209 through steps 203 to 204 described herein. Here, the reason why the time conversion process is performed in the main routine without setting the NE interrupt within the injection angle and the remainder angle will be described. In the present invention, when the correction process is performed within the NE interrupt, the time T0 required for the correction is set.
On the other hand, it is determined whether the valve opening time T is large. However, even in the absence of this correction, from the interruption of the NE to the setting of the valve opening time T, that is, T ′ ≦ T0 in FIG.
The time Ta up to Step 209 corresponding to the time (about 80
μs). Therefore, when the valve opening time T ′ <Ta,
At this time, the injection angle k is too late. In this process, the injection angle is set to k ′ = k−1, the valve opening time T ′ = T + interval time of the injection angle k, and the injection angle is set from the immediately preceding injection angle. That is, as described in FIG.
When T <Ta calculated by T45 before CA, the injection angle 4
Is set to −1 to make the injection angle = 3, and the valve opening time is accordingly set to T and T
34, the NE interrupt, the rotation angle No. At 3, the valve opening time T '= T + T34 is set. At this time, of course, the present invention goes through a routine for performing a correction process. That is, when T ′ <Ta, it is understood that the processing in the NE interrupt of the injection angle is slow because the injection angle needs to be set to −1. This processing is the main routine in the present embodiment. Alternatively, the processing may be synchronized every time the accelerator opening or the engine speed data is updated, or further, may be synchronized with a specific rotation angle in time for the injection angle k. Therefore, in order to increase the control accuracy, it is necessary to use a microcomputer having a high processing speed and to perform programming by giving priority to the setting process of the valve opening time T within the NE interrupt so as to reduce Ta. Also. Even in the correction process of the present invention, the time T0 required for the correction is short, and by performing the correction accurately, the process with less deviation can be executed. As described above, according to the present invention, the step generated at the time of switching the acceleration / deceleration injection angle in order to convert the remaining angle time from the rotation angle pulse interval before 180 ° CA when the remaining angle time converted value T ′ is small. Since the error is small, and when it is large, there is ample time to correct the error. Therefore, within the NE interrupt, the correction is performed based on the difference between the immediately preceding rotation angle time interval and the time interval of the same rotation angle before 180 ° CA. Even when switching the injection angle, the step is small,
It is characterized by a smooth injection change and no effect on rotation fluctuation.

[0025]

As described above, according to the present invention, the present time interval Tk-1, k and the previous time interval T'k-
Find the difference from 1, k, further find the correction coefficient from this difference,
The correction coefficient is multiplied by the time T 'to correct the current time T, and the correction processing is performed. However, since the correction processing time T0 is required, it is first determined whether the time T' is equal to or longer than the correction processing time T0. In such a case, the above correction is performed on the time T ′, and a time T that is close to true can be predicted in consideration of acceleration or deceleration. On the other hand, if the time T 'is less than the correction processing time T0, the effect of acceleration or deceleration is small and there is no need for correction, so the uncorrected time T' is used directly in the calculation of the injection end time. Thus, as in the conventional case where there is acceleration or deceleration, a step in the injection amount caused when the injection angle is changed due to a change in rotation can be reduced, and the connection can be made smoothly.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a solenoid valve spill system based on a Bosch distribution type fuel injection pump according to an embodiment of the present invention.

FIG. 2 is a diagram showing a rotation angle sensor 50 in a section aa of FIG. 1;

FIG. 3 is a timing chart showing a basic concept of injection amount adjustment by an electronic control unit.

FIG. 4 is a view showing a rotation angle pulse signal and an opening signal of a spill adjustment solenoid valve according to movements during stable rotation and acceleration and deceleration in the viewpoint of the present invention.

FIG. 5 is a diagram showing an injection amount with respect to a valve opening command angle of a spill adjustment solenoid valve with respect to the viewpoint of the present invention.

FIG. 6 is a flowchart for explaining energization opening control of the electromagnetic valve 21 by the electronic control device 26 of FIG. 1;

FIG. 7 is a diagram illustrating a correction coefficient at time T ′ obtained by converting the remainder angle θh into angle time.

FIG. 8 is a time chart for explaining a conventional solenoid valve opening / closing timing.

FIG. 9 is a diagram showing a relationship between an injection amount and a valve opening command angle during acceleration or deceleration.

Claims (1)

(57) [Claims] A spill angle corresponding to the injection amount is output for each crank angle to control the fuel injection amount by overflowing high-pressure fuel by opening a solenoid valve at a desired timing. A fuel injection control device that obtains the rotation angle pulse number k, which is a quotient of an integer divided by an angle, and the remainder angle, converts the remainder angle into time, and forms an injection end signal, uses linear interpolation from the previous remainder angle. A first memory for storing a time T ′ converted to a time by the following, and a time interval Tk−1, k of the rotation angle pulse (K) immediately before the current injection.
And a time interval T'k-1, k of the rotation angle pulse (K) immediately before the previous injection.
A third memory for storing the current time interval Tk-1, k and the previous time interval T'k
A fourth memory for storing a correction coefficient for correcting the time T using the difference between the first time and the first time as a parameter, and a fourth processing time for storing a correction processing time T0 which is a maximum time required for correcting the time T ′. And when the time T 'is equal to or longer than the correction processing time T0, the difference between the present time interval Tk-1, k and the previous time interval T'k-1, k is obtained. When the correction coefficient is further obtained from the difference, the correction coefficient is multiplied by the time T 'to correct the current time T and the injection end time is formed from the current time T. On the other hand, the time T' is shorter than the correction processing time T0. A fuel injection control device, wherein an injection end time is formed from a time T ′ that is not corrected to a predetermined time.