JP7000720B2 - Fuel injection control device for internal combustion engine - Google Patents

Description

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

As a fuel injection valve that injects and supplies fuel to each cylinder of an internal combustion engine mounted on a vehicle or the like, for example, an electromagnetic solenoid type is known. In this type of fuel injection valve, the fuel injection timing and fuel injection are performed by controlling the energization timing and energization time of the coil built in the fuel injection valve body to drive the valve body (needle) in the valve opening direction. The amount is controlled.

In order to improve engine output, fuel efficiency, exhaust emissions, etc., it is necessary to accurately control the fuel injection amount. Therefore, various techniques for correcting injection amount variations have been conventionally proposed (see, for example, Patent Document 1). Patent Document 1 discloses that the magnetic circuit is energized with a predetermined energization pulse in a range in which the valve body of the fuel injection valve is not driven, and the magnetic circuit variation value is calculated from the negative terminal voltage of the fuel injection valve at that time. Has been done.

Japanese Unexamined Patent Publication No. 2016-205276

It is conceivable that the cause of the variation in the injection amount is not only the influence of the individual difference on the fuel injection valve side but also the influence of the individual difference on the control unit side that controls the drive of the fuel injection valve. In order to control the fuel injection amount with high accuracy, it is necessary to detect the injection amount variation with high accuracy.

The present invention has been made in view of the above problems, and a main object thereof is to provide a fuel injection control device for an internal combustion engine capable of performing accurate injection amount correction.

Hereinafter, means for solving the above problems and their actions and effects will be described. In the following, for the sake of easy understanding, the reference numerals of the corresponding configurations in the embodiments of the invention are appropriately shown in parentheses or the like, but the present invention is not limited to the specific configurations shown in the parentheses or the like.

The present invention relates to a fuel injection control device for an internal combustion engine applied to a fuel injection system of an internal combustion engine (10) including a fuel injection valve (30) and control units (41, 42) for driving the fuel injection valve. .. The invention according to claim 1 carries out the first energization in which the valve body (34) of the fuel injection valve is driven to energize the fuel injection valve, and the fuel after the energization of the first energization is completed. A first information acquisition unit that acquires first drive information as information on the drive characteristics of the fuel injection valve based on the terminal voltage of the injection valve, and a second that energizes the fuel injection valve in a manner in which the valve body is not driven. The second information acquisition unit and the first information acquisition unit acquire the second drive information as information on the drive characteristics based on the terminal voltage after the energization of the second energization is completed. It is provided with an injection amount calculation unit that calculates an injection correction amount of the fuel injection valve based on the first drive information obtained and the second drive information acquired by the second information acquisition unit.

In the above configuration, by using the terminal voltage of the fuel injection valve after the end of the first energization, the injection amount variation due to the lift behavior of the fuel injection valve and the control unit are used as information on the drive characteristics of the fuel injection valve. Information (first drive information) including the injection amount variation caused by the circuit element of the above can be acquired. Further, by using the terminal voltage of the fuel injection valve after the end of the energization of the second energization, the information on the drive characteristics of the fuel injection valve does not include the variation in the injection amount due to the lift behavior of the fuel injection valve, and the control unit can be used. Information (second drive information) including injection amount variation due to the circuit element can be acquired. Therefore, according to the first drive information and the second drive information, it is possible to separate the injection amount variation caused by the lift behavior of the fuel injection valve and the injection amount variation caused by the circuit element of the control unit. As a result, it is possible to accurately grasp the individual injection amount variations of the fuel injection valve and the control unit, and it is possible to accurately correct the injection amount.

Schematic block diagram of the engine control system. It is a figure which shows the drive state of a fuel injection valve, (a) is a figure which shows the full lift state, (b) is a figure which shows the partial lift state. The figure which shows the partial lift area and the full lift area. The figure which shows the behavior when fuel is injected from a fuel injection valve. A time chart showing the transition of the negative terminal voltage and the threshold value arrival time after the injection pulse is turned off. The figure which shows the difference of the waveform of the flyback voltage due to the mechanical variation and the electrical variation. Functional block diagram of fuel injection control. Functional block diagram of fuel injection control. The flowchart which shows the processing procedure of the variation value calculation processing. The flowchart which shows the processing procedure of the calculation process of the threshold value arrival time. A flowchart showing the processing procedure of the zero lift learning process. A time chart showing a specific mode when switching from the normal mode to the zero lift learning mode. A time chart showing a specific mode when switching from the zero lift learning mode to the normal mode.

Hereinafter, embodiments will be described with reference to the drawings. This embodiment is embodied as a control system for controlling an in-cylinder injection type multi-cylinder gasoline engine for a vehicle. First, a schematic configuration of an engine control system will be described with reference to FIG. In each of the following embodiments, the parts that are the same or equal to each other are designated by the same reference numerals, and the description thereof will be used for the parts having the same reference numerals.

In the engine 11, an air cleaner 13 is provided at the most upstream portion of the intake pipe 12, and an air flow meter 14 for detecting the intake air amount is provided on the downstream side of the air cleaner 13. On the downstream side of the air flow meter 14, a throttle valve 16 whose opening degree is adjusted by a motor 15 and a throttle opening degree sensor 17 for detecting the opening degree (throttle opening degree) of the throttle valve 16 are provided.

A surge tank 18 is provided on the downstream side of the throttle valve 16, and an intake pipe pressure sensor 19 for detecting the intake pipe pressure is provided in the surge tank 18. An intake manifold 20 for introducing air into each cylinder 21 of the engine 11 is connected to the surge tank 18. Each cylinder 21 of the engine 11 is provided with an electromagnetic fuel injection valve 30 that directly injects fuel into the cylinder. A spark plug 22 is attached to the cylinder head of the engine 11 for each cylinder 21, and the air-fuel mixture in the cylinder is ignited by the spark discharge of the spark plug 22 of each cylinder 21.

The exhaust pipe 23 of the engine 11 is provided with an exhaust sensor 24 that detects the air-fuel ratio of the air-fuel mixture, rich / lean, or the like by targeting the exhaust gas. The exhaust sensor 24 is an air-fuel ratio sensor, an oxygen sensor, or the like. A catalyst 25 such as a three-way catalyst for purifying the exhaust gas is provided on the downstream side of the exhaust sensor 24.

A cooling water temperature sensor 26 for detecting the cooling water temperature and a knock sensor 27 for detecting knocking are attached to the cylinder block of the engine 11. A crank angle sensor 29 that outputs a pulse signal each time the crank shaft 28 rotates by a predetermined crank angle is attached to the outer peripheral side of the crank shaft 28. The crank angle and the engine rotation speed are detected based on the crank angle signal of the crank angle sensor 29. The outputs of these various sensors are input to the ECU 40.

The ECU 40 is an electronic control unit mainly composed of a microcomputer, and uses a control program stored in a built-in ROM (storage medium) to perform various controls of the engine 11 based on detection signals of various sensors. do. The ECU 40 corresponds to the fuel injection control device. The ECU 40 calculates a required injection amount according to the engine operating state, controls the drive of the fuel injection valve 30 based on the required injection amount, and controls the ignition timing of the spark plug 22.

More specifically, the ECU 40 includes a microcomputer 41 for engine control that performs fuel injection control and a drive IC 42 for driving a fuel injection valve. The microcomputer 41 calculates the required injection amount based on the engine operating state, and also calculates the injection pulse width based on the calculated required injection amount. Further, the calculated injection pulse width is output to the drive IC 42. The drive IC 42 drives the fuel injection valve 30 by the injection pulse generated based on the injection pulse width. As a result, the fuel for the required injection amount is injected from the fuel injection valve 30.

In the present embodiment, the engine 11 is a 4-cylinder gasoline engine, and as fuel injection by the fuel injection valve 30 provided in each cylinder, injection pulses are output in each of the intake stroke and the compression stroke in one combustion cycle to generate fuel. Is injected. The combustion order of the first cylinder # 1 to the fourth cylinder # 4 is # 1 → # 3 → # 4 → # 2. At the time of fuel injection, two cylinders having a combustion order of every other are grouped into drive groups 1 and 2, and each fuel injection valve 30 is driven by a drive system for each drive group. In the present embodiment, the first cylinder # 1 and the fourth cylinder # 4 belong to the drive group 1, and the second cylinder # 2 and the third cylinder # 3 belong to the drive group 2.

The fuel injection valve 30 is provided with a voltage sensor 43 for detecting the negative terminal voltage Vm and a current sensor 44 for detecting the energization current flowing through the electromagnetic unit (coil). The detection results of the voltage sensor 43 and the current sensor 44 are sequentially input to the ECU 40.

In the fuel injection control of the present embodiment, the full lift injection in which the valve body of the fuel injection valve 30 reaches the full lift position and the desired amount of fuel is injected in the full lift state, and the valve body of the fuel injection valve 30 is in the full lift position. In the partial lift state before reaching, the movement of the valve body to the valve opening side is completed, and in that state, the partial lift injection in which a desired minute amount of fuel is injected is carried out. The partial lift injection is performed, for example, in the fuel injection (compression stroke injection in the present embodiment) performed after the main injection (intake stroke injection in the present embodiment).

Full lift injection and partial lift injection will be described with reference to FIG. In FIG. 2, (a) shows the operation at the time of full lift injection, and (b) shows the operation at the time of partial lift injection.

As shown in FIG. 2, the fuel injection valve 30 has a coil 31 as an electromagnetic part that generates an electromagnetic force by energization, a fixed core 32 made of a magnetic material, and a fixed core 32 made of a magnetic material on the side of the fixed core 32 by the electromagnetic force. The movable core 33 to be sucked, the needle-shaped valve body 34 driven integrally with the movable core 33, the first spring 35 for urging the valve body 34 to the valve closing side, and the movable core 33 to be counter-closed. It has a second spring 36 that urges the side. When the coil 31 is energized, the valve body 34 moves away from the valve seat to the valve opening side, so that the fuel injection valve 30 is in the valve opening state and fuel injection is performed. The urging force of the second spring 36 is set to be smaller than the urging force of the first spring 35.

In FIGS. 2A and 2B, the injection pulse widths are different. As shown in FIG. 2A, when the valve body lift amount becomes the full lift amount, the injection pulse width becomes relatively long. In this case, the valve body 34 reaches the full lift position where the movable core 33 abuts on the stopper 32a on the fixed core 32 side. On the other hand, as shown in FIG. 2B, when the valve body lift amount becomes the partial lift amount, the injection pulse width becomes relatively short. In this case, the valve body 34 is in a state before the movable core 33 abuts on the stopper 32a, and is in a partial lift state in which the full lift position is not reached. When the energization of the coil 31 is stopped due to the fall of the injection pulse, the movable core 33 and the valve body 34 return to the valve closed position, so that the fuel injection valve 30 is closed and the fuel injection is stopped. Since the movable core 33 and the valve body 34 are formed separately, when the valve body 34 reaches the closed position, the valve body 34 is held in the closed position, whereas the movable core 33 moves to the tip side by itself.

FIG. 3 is a diagram showing a partial lift region in which partial lift injection is performed and a full lift region in which full lift injection is performed. As shown in FIG. 3, in both the partial lift region and the full lift region, the longer the injection pulse width (that is, the energization time), the more the fuel injection amount tends to increase linearly.

Here, in the in-cylinder injection engine 11 that injects high-pressure fuel into the cylinder from the fuel injection valve 30, the linearity of the change characteristic of the fuel injection amount with respect to the injection pulse width tends to deteriorate in the partial lift region. , The individual difference in the fuel injection amount corresponding to the injection pulse width tends to be large. On the other hand, if the injection amount variation is large, the fuel amount injected from the fuel injection valve 30 cannot be precisely adjusted, and there is a concern that exhaust emissions, drivability, and the like may be deteriorated. In particular, in the partial lift injection, the injection amount is very small, and the influence of the injection amount variation tends to be large due to this minute injection. Therefore, it is necessary to accurately detect individual injection amount variations in each injection system of each fuel injection valve 30 and ECU 40, and to control the injection amount according to the injection amount variations.

The variation in the injection amount of the fuel injection valve 30 includes the variation due to individual differences in the lift behavior of the fuel injection valve 30 (hereinafter, also referred to as “mechanical variation”) and the microcomputer 41 that controls the drive of the fuel injection valve 30. There are variations due to individual differences in the circuit elements of the drive IC 42 (hereinafter, also referred to as “electrical variations”). The mechanical variation is the variation in the injection amount on the fuel injection valve 30 side, which is an actuator that performs fuel injection, and is caused by, for example, the assembled state of the fixed core 32 and the movable core 33, the ease of movement of the valve body 34, and the like. Occurs. On the other hand, the electrical variation is an injection amount variation on the ECU 40 side that controls the drive of the fuel injection valve 30, and is caused by, for example, magnetic efficiency in the circuit elements of the microcomputer 41 and the drive IC 42. In this case, the larger the magnetic loss in the circuit element of the ECU 40 (that is, the lower the magnetic efficiency), the shorter the time required from turning off the injection pulse to the fuel injection valve 30 to closing the valve, and the valve closing time. Will be faster.

Therefore, in the present embodiment, among the injection amount variations of the fuel injection valve 30, the electric variation value, which is the injection correction amount for correcting the electrical variation, and the machine, which is the injection correction amount for correcting the mechanical variation. The target variation value is calculated, and the correction amount (injection correction amount) of the fuel injection amount is calculated based on the calculated electrical variation value and mechanical variation value. The mechanical variation corresponds to the first correction amount, and the electrical variation value corresponds to the second correction amount. The calculation process of the injection amount variation of the present embodiment will be described in detail below.

In the fuel injection valve 30, the negative terminal voltage Vm changes due to the dielectric electromotive force as the injection pulse is turned off. This characteristic will be described with reference to FIG. FIG. 4 shows the behavior when a predetermined amount of fuel is injected from the fuel injection valve 30 by driving the valve body 34 of the fuel injection valve 30. In FIG. 4, (a) is for the positive terminal voltage Vp of the fuel injection valve 30, (b) is for the negative terminal voltage Vm, (c) is for the injector drive current, and (d) is the injection rate of the fuel injection valve 30. The behavior of each is shown. As can be seen from FIG. 4, when the coil 31 of the fuel injection valve 30 is energized to open the valve and then the energization is stopped after the lapse of time corresponding to the injection pulse width, the battery voltage is applied for a predetermined period immediately after the energization is turned off. A flyback voltage (counter electromotive voltage) higher than VB is generated (see FIG. 4 (b)).

In this specification, the magnitude of the negative terminal voltage Vm is expressed as an absolute value. For example, "Vm is larger than a predetermined value" means that Vm is a value larger than the predetermined value on the negative side.

By using such a flyback voltage, it is possible to detect the actual valve closing timing of the fuel injection valve 30 (see FIG. 5). Specifically, the ECU 40 uses the negative terminal voltage Vm after turning off the injection pulse as the cutoff frequency at the first frequency f1, which is lower than the frequency of the noise component, during the fuel injection period by the partial lift injection. The first filter voltage Vsm1 filtered by the first low-pass filter (smoothing process) is calculated. The first low-pass filter is a low-pass filter whose pass region is a frequency band lower than the cutoff frequency (first frequency f1).

Further, the ECU 40 filters the negative terminal voltage Vm after turning off the injection pulse with a second low-pass filter having a second frequency f2 lower than the first frequency f1 as a cutoff frequency. The filter voltage Vsm2 is calculated. The filter processing using the first low-pass filter and the filter processing using the second low-pass filter are performed for each cylinder of the engine 11. By such processing, the ECU 40 calculates the first filter voltage Vsm1 from which the noise component is removed from the negative terminal voltage Vm and the second filter voltage Vsm2 for detecting the valve closing position. The second low-pass filter is a low-pass filter whose pass region is a frequency band lower than the cutoff frequency (second frequency f2).

Subsequently, the difference Vdiff (= Vsm1-Vsm2) between the first filter voltage Vsm1 and the second filter voltage Vsm2 is calculated. Further, the time required from the reference timing after the injection pulse is turned off (t3 in FIG. 5) to the timing when the difference Vdiff exceeds the threshold value Vt (t4 in FIG. 5) is calculated, and this is defined as the threshold value arrival time Tdiff. In the present embodiment, paying attention to the fact that there is a correlation between the difference Vdiff and the valve closing timing, the timing when the difference Vdiff exceeds the threshold value Vt is detected as the valve closing timing. The threshold value arrival time Tdiff is calculated for each cylinder of the engine 11 and the valve closing timing is calculated for each cylinder. The threshold arrival time Tdiff corresponds to "information regarding the drive characteristics of the fuel injection valve 30".

Reference timing In this embodiment, the reference timing is the timing at which the negative terminal voltage Vm falls below the determination value Voff after the injection pulse is turned off (in other words, the first timing at which the negative terminal voltage Vm shifts from larger to smaller than the determination value Voff). .. The reference timing may be set to when the injection pulse is on (t1 in FIG. 5) or when the injection pulse is off (t2 in FIG. 5). The threshold value Vt may be a variable value according to the fuel pressure, the fuel temperature, or the like, or may be a preset fixed value.

Further, in the present embodiment, the flyback voltage generated when the energization of the coil 31 is turned off is used, and the mechanical variation value and the electrical variation value are calculated based on the behavior of the negative terminal voltage Vm after the injection pulse is turned off. is doing. Specifically, the ECU 40 performs lift learning to calculate the threshold value arrival time Tdiff (corresponding to the first drive information) based on the negative terminal voltage Vm after the energization is turned off when the partial lift injection is performed. Further, the ECU 40 energizes the fuel injection valve 30 with a predetermined injection pulse within a range in which the valve body 34 is not driven during the period when fuel is not injected from the fuel injection valve 30 (in the present embodiment, fuel is being cut). Zero lift learning is performed to calculate the threshold arrival time Tdiff (corresponding to the second drive information) based on the negative terminal voltage Vm after the energization is turned off. Then, the threshold arrival time Tdiff (hereinafter, also referred to as “threshold arrival time Tdiff_1”) calculated by lift learning and the threshold arrival time Tdiff (hereinafter, also referred to as “threshold arrival time Tdiff_1”) calculated by zero lift learning. The mechanical variation value and the electrical variation value are calculated based on. The electrical variation value corresponds to the "first correction amount", and the mechanical variation value corresponds to the "second correction amount".

FIG. 6 shows the difference in the waveform of the flyback voltage due to the mechanical variation and the electrical variation. In FIG. 6, (a) shows the time of lift learning, and (b) shows the time of zero lift learning. In FIG. 6, it is assumed that the magnetic efficiency of the circuit element of the ECU 40 is worse than that of the nominal, and the flyback voltage converges quickly.

In the zero lift learning, the coil 31 is energized, but the valve body 34 is not driven. In this case, the negative terminal voltage Vm after the injection pulse is turned off is not affected by the mechanical variation, but is affected by the electrical variation and deviates from the nominal value. That is, as shown in FIG. 6B, the amount of deviation from the nominal value of the negative terminal voltage Vm (S1 in FIG. 6) when energization is performed by the injection pulse in the range in which the valve body 34 is not driven is electrical. It can be considered to be due to variability.

On the other hand, in lift learning, the valve body 34 is driven by energizing the coil 31, and fuel is injected from the fuel injection valve 30. In this case, the negative terminal voltage Vm after the injection pulse is turned off is affected by both mechanical variation and electrical variation and deviates from the nominal value. That is, as shown in FIG. 6A, the amount of deviation from the nominal value of the negative terminal voltage Vm when energized by the injection pulse in the range driven by the valve body 34 is the amount of deviation due to electrical variation. It can be regarded as the sum of (S1 in FIG. 6) and the amount of deviation due to mechanical variation (S2 in FIG. 6). Therefore, the deviation amount due to the mechanical variation can be obtained except for the deviation amount S1 minute in the zero lift learning.

The mechanical variation is a value determined for each fuel injection valve 30, and the electrical variation is a value determined for each drive system of the ECU 40. According to the lift learning and the zero lift learning, it is possible to acquire the mechanical variation for each fuel injection valve 30, and also to acquire the electrical variation for each drive system of the ECU 40.

7 and 8 are functional block diagrams of the fuel injection control of the present embodiment. As shown in FIG. 7, the microcomputer 41 of the ECU 40 includes an injection pulse calculation unit 51 and a storage unit 52. The injection pulse calculation unit 51 calculates the basic injection amount based on the engine operating state (for example, engine rotation speed, engine load, etc.). Further, the required injection amount is calculated by reading out the correction parameter stored in the storage unit 52 and correcting the basic injection amount based on the read-out correction parameter. In this embodiment, the correction parameter includes a mechanical variation value and an electrical variation value. The injection pulse calculation unit 51 generates a signal representing the injection pulse width Ti according to the required injection amount, and outputs the signal to the drive IC 42.

The drive IC 42 includes a drive control unit 53 and a calculation unit 54. The drive control unit 53 inputs a signal representing the injection pulse width Ti from the injection pulse calculation unit 51, and opens the fuel injection valve 30 by the injection pulse generated based on the injection pulse width Ti. The calculation unit 54 includes a zero lift learning unit 55 and a lift learning unit 56 as means for calculating a correction amount (deviation amount) of the fuel injection amount based on the behavior of the negative terminal voltage Vm after the injection pulse is turned off. There is.

As shown in FIG. 8, the zero lift learning unit 55 includes a Tdiff calculation unit 55A, a reference value storage unit 55B, a correction amount calculation unit 55C, and a correction amount storage unit 55D. When the predetermined zero lift learning execution condition is satisfied, the Tdiff calculation unit 55A energizes the fuel injection valve 30 by an injection pulse in a range in which the valve body 34 of the fuel injection valve 30 is not driven, and calculates the threshold value arrival time Tdiff_2. The correction amount calculation unit 55C inputs the threshold value arrival time Tdiff_2 from the Tdiff calculation unit 55A, and acquires a reference value (nominal value, constant) from the reference value storage unit 55B. Then, the electrical variation value Q1 is calculated as the deviation amount of the threshold value arrival time Tdiff_2 with respect to the reference value. The electrical variation value Q1 is represented by, for example, the following equation (1).
Q1 = (threshold arrival time Tdiff_2) / reference value ... (1)
The calculated electrical variation value Q1 is stored in the correction amount storage unit 55D.

As shown in FIG. 8, the lift learning unit 56 includes a Tdiff calculation unit 56A, a first calculation unit 56B, an estimated injection amount calculation unit 56C, a reference injection amount storage unit 56D, and a second calculation unit 56E. ing. When a predetermined lift learning execution condition is satisfied, the Tdiff calculation unit 56A energizes the fuel injection valve 30 by an injection pulse that can displace the valve body 34 to a predetermined partial lift position, and calculates the threshold value arrival time Tdiff_1. The first calculation unit 56B inputs the threshold value arrival time Tdiff_1 from the Tdiff calculation unit 56A, and acquires the electrical variation value Q1 from the correction amount storage unit 55D. Then, the threshold arrival time Tdiff_1 is multiplied by the electrical variation value Q1 to calculate the corrected threshold arrival time Tdiff * Q1. The corrected threshold value arrival time Tdiff * Q1 is information on the valve closing timing in which the electrical variation is removed from the threshold arrival time Tdiff_1 and the mechanical variation is included.

The estimated injection amount calculation unit 56C calculates the estimated injection amount R from the corrected threshold value arrival time Tdiff * Q1. Here, the estimated injection amount R is calculated by multiplying the corrected threshold value arrival time Tdiff * Q1 by the correction coefficient K. The second calculation unit 56E inputs the estimated injection amount R from the estimated injection amount calculation unit 56C, and acquires the reference injection amount (nominal value, constant) from the reference injection amount storage unit 56D. Then, the mechanical variation value Q2 is calculated as the deviation amount of the estimated injection amount R with respect to the reference injection amount. The mechanical variation value Q2 is represented by, for example, the following equation (2).
Q2 = Estimated injection amount / Reference injection amount ... (2)

The microcomputer 41 of the ECU 40 stores the electrical variation value Q1 and the mechanical variation value Q2 input from the drive IC 42 in the storage unit 52 as a correction amount of the injection amount variation. The microcomputer 41 corrects the basic injection amount using the electric variation value Q1 and the mechanical variation value Q2 stored in the storage unit 52, and opens and drives the fuel injection valve 30 based on the corrected injection pulse. .. The lift learning unit 56 realizes the function of the "first information acquisition unit", and the zero lift learning unit 55 realizes the function of the "second information acquisition unit". Further, the lift learning unit 56 and the zero lift learning unit 55 realize the function of the "correction amount calculation unit".

When calculating the electrical variation value Q1, the threshold arrival time Tdiff_2 is acquired by energizing once for zero lift learning, and the electrical variation value Q1 is calculated using the acquired threshold arrival time Tdiff_2. You may. Alternatively, a plurality of threshold arrival times Tdiff_2 may be acquired by energizing for zero lift learning a plurality of times, and the electrical variation value Q1 may be calculated from the average value thereof. Similarly, for the mechanical variation value Q2, the mechanical variation value Q2 may be calculated by one threshold arrival time Tdiff_1, and the mechanical variation is mechanically varied from the average value of the threshold arrival time Tdiff_1 acquired by a plurality of fuel injections. The value Q2 may be calculated.

Next, the processing procedure of the variation value calculation processing in the present embodiment will be described with reference to the flowchart of FIG. This process is executed by the drive IC 42 every predetermined cycle (for example, every one combustion cycle).

In FIG. 9, in step S101, it is determined whether or not the zero lift learning is completed. If the zero lift learning has not been completed yet, the process proceeds to step S102, and it is determined whether or not the zero lift learning execution condition is satisfied.

The following conditions B1 to B7 are set as the zero lift learning execution conditions, and when all of these conditions B1 to B7 are satisfied, a positive determination is made in step S102.
B1: Zero lift learning has not been completed in this driving cycle.
B2: Fuel is being cut during deceleration for all cylinders.
B3: The engine rotation speed is equal to or higher than the predetermined return rotation speed for returning from the fuel cut state to the combustion state.
B4: The pressure (fuel pressure) of the fuel injected from the fuel injection valve 30 is equal to or higher than the predetermined fuel pressure that can avoid unintended fuel injection due to energization for zero lift learning.
B5: The battery voltage is equal to or higher than the predetermined voltage that can guarantee the operation of the fuel injection valve 30.
B6: The pressure inside the cylinder is equal to or higher than the predetermined pressure that can avoid unintended fuel injection due to energization for zero lift learning.
B7: The fail-safe is not running.

If the zero lift learning execution condition is not satisfied, this routine is terminated as it is. On the other hand, if the zero lift learning execution condition is satisfied, the process proceeds to step S103, and zero lift learning is executed according to the flowchart shown in FIG. After that, in step S104, the electrical variation value Q1 calculated by zero lift learning is output to the microcomputer 41, and is stored or updated in the storage unit 52 of the microcomputer 41.

If the zero lift learning is completed, a positive determination is made in step S101 and the process proceeds to step S105. In step S105, it is determined whether or not the lift learning execution condition is satisfied. The lift learning execution conditions include at least the conditions that zero lift learning has been completed, that fuel is not being cut, and that the catalyst is being warmed up early, and that all of these conditions are satisfied. It is judged affirmatively.

If the lift learning execution condition is satisfied, the process proceeds to step S106 to execute the lift learning. In step S107, the mechanical variation value Q2 calculated by lift learning is output to the microcomputer 41, and is stored or updated in the storage unit 52 of the microcomputer 41.

Next, the process of calculating the threshold value arrival time Tdiff in the lift learning and the zero lift learning will be described with reference to the flowchart of FIG. This process is executed by the drive IC 42 at predetermined cycles (for example, every several msec).

In FIG. 10, in step S201, the negative terminal voltage Vm of the fuel injection valve 30 is acquired. In the following step S202, the negative terminal voltage Vm is filtered by the first low-pass filter to calculate the first filter voltage Vsm1. The first low-pass filter is a digital filter that obtains the current value Vsm1 (k) of the first filter voltage by using the previous value Vsm1 (k-1) of the first filter voltage and the current value Vm (k) of the negative terminal voltage. It is implemented by the following equation (3).
Vsm1 (k) =
{(N1-1) / n1} x Vsm1 (k-1) + (1 / n1) x Vm (k) ... (3)

The time constant n1 of the first low-pass filter is the following (4) using the sampling frequency fs (= 1 / Ts) of the negative terminal voltage Vm and the cutoff frequency (first frequency f1) of the first low-pass filter. It is set to satisfy the relation of expressions.
1 / fs: 1 / f1 = 1: (n1-1) ... (4)

In the following step S203, the second filter voltage Vsm2 obtained by filtering the negative terminal voltage Vm with the second low-pass filter having the second frequency f2 lower than the first frequency f1 as the cutoff frequency is calculated. The second low-pass filter is a digital filter that obtains the current value Vsm2 (k) of the second filter voltage by using the previous value Vsm2 (k-1) of the second filter voltage and the current value Vm (k) of the negative terminal voltage. It is implemented by the following equation (5).
Vsm2 (k) =
{(N2-1) / n2} x Vsm2 (k-1) + (1 / n2) x Vm (k) ... (5)
The time constant n2 of the second low-pass filter is the following (6) using the sampling frequency fs (= 1 / Ts) of the negative terminal voltage Vm and the cutoff frequency (second frequency f2) of the second low-pass filter. It is set to satisfy the relation of expressions.
1 / fs: 1 / f2 = 1: (n2-1) ... (6)

In the following step S204, the difference Vdiff (= Vsm1-Vsm2) between the first filter voltage Vsm1 and the second filter voltage Vsm2 is calculated. It should be noted that the guard processing may be performed so that the difference Vdiff does not become 0 or more, and only the negative component may be extracted.

In the following step S205, the threshold value Vt is acquired and the previous value Tdiff (k-1) of the threshold value arrival time is acquired. In step S206, it is determined whether or not the injection pulse is off. If an affirmative determination is made in step S206, the process proceeds to step S207, and whether or not the negative terminal voltage Vm of the fuel injection valve 30 is lower than the determination value Voff (timing from larger to smaller than the determination value Voff). Is determined. If an affirmative determination is made in step S207, the process proceeds to step S209, and the current value Tdiff (k) of the voltage inflection point time is reset to "0".

On the other hand, if a negative determination is made in step S207, the process proceeds to step S208, and whether or not the difference Vdiff between the first filter voltage Vsm1 and the second filter voltage Vsm2 exceeds the threshold value Vt (from smaller to larger than the threshold value Vt). Whether or not it has become) is determined. If a negative determination is made in step S208, the process proceeds to step S210, the predetermined value Ts (calculation cycle of this routine) is added to the previous value Tdiff (k-1) of the threshold arrival time, and the current value Tdiff of the threshold arrival time. By obtaining (k), the threshold arrival time Tdiff is counted up.

On the other hand, if it is determined in step S208 that the difference Vdiff exceeds the threshold value Vt, the process proceeds to step S211 and the current value Tdiff (k) of the threshold value arrival time is held at the previous value Tdiff (k-1). As a result, the time from the timing (reference timing) when the negative terminal voltage Vm of the fuel injection valve 30 falls below the determination value Voff after the injection pulse is turned off to the timing when the differential Vdiff exceeds the threshold value Vt is calculated as the threshold value arrival time Tdiff.

If it is determined in step S206 that the injection pulse is not off (that is, the injection pulse is on), the current value Tdiff (k) is held at the previous value Tdiff (k-1) and the threshold value is reached. The calculated value of the time Tdiff is held until the next reference timing.

Next, the processing procedure of zero lift learning will be described with reference to the flowchart of FIG. This process is executed by the drive IC 42 at a predetermined calculation cycle (for example, at each injection update timing).

In FIG. 11, in step S301, it is determined whether or not the zero lift learning execution conditions (B1 to B7) are satisfied. If the zero lift learning execution condition is satisfied, the process proceeds to step S302, the execution condition establishment flag is set, and the injection mode for zero lift learning is set. Specifically, the fuel injection in each cylinder is set to one intake stroke. Further, the injection timing is set to a predetermined time for zero lift learning, and the injection time is set to a predetermined time Doff for zero lift learning. The predetermined time Doff is set to a value corresponding to the energization pulse width in which the valve body 34 of the fuel injection valve 30 is not driven (see FIG. 6B).

In step S303, the zero lift learning counter is counted up. In the following step S304, it is determined whether or not the zero lift learning counter is larger than the first determination value TH1. If the zero lift learning counter is equal to or less than the first determination value TH1, this routine is temporarily terminated. On the other hand, when the zero lift learning counter is larger than the first determination value TH1, the process proceeds to step S305, the injector drive current is set to the predetermined current value Aoff for zero lift learning, and the fuel injection valve 30 is driven during fuel cut. Allow. Here, the predetermined current value Off is a current value at which the valve body 34 of the fuel injection valve 30 is not driven. The predetermined current value Aoff may be set variably according to the engine operating state (for example, fuel pressure, etc.), or may be a fixed value.

In the following step S306, it is determined whether or not the zero lift learning counter is larger than the second determination value TH2. The process proceeds to step S307 on condition that the zero lift learning counter is larger than the determination value TH2. In step S307, the detection of the zero lift learning value (in the present embodiment, the threshold arrival time Tdiff_2) is permitted. Further, in step S308, the electrical variation value Q1 is calculated based on the threshold arrival time Tdiff_2.

If the zero lift learning execution condition is not satisfied, the process proceeds to step S309, and the injection mode is set to the normal mode. Further, in step S310, the zero lift learning counter is reset and the detection of the zero lift learning value is prohibited. After that, this routine ends.

According to the flowchart of FIG. 11, the ECU 40 energizes the injection mode from the normal mode, which is the fuel injection mode at the normal time, in a range in which the valve body 34 of the fuel injection valve 30 is not driven, and acquires the threshold arrival time Tdiff. When switching to the zero lift learning mode, the zero lift learning mode is not immediately switched to the zero lift learning mode even if the zero lift learning execution condition is satisfied, but is switched to the zero lift learning mode after a predetermined delay time has elapsed. On the other hand, when switching from the zero lift learning mode to the normal mode, the mode is immediately switched to the normal mode when the zero lift learning execution condition is not satisfied. This control will be described with reference to the time charts of FIGS. 12 and 13. The black triangle marks in FIGS. 12 and 13 indicate the injection update timing.

First, the time of switching from the normal mode to the zero lift learning mode will be described with reference to FIG. In FIG. 12, consider a case where the fuel cut is started at the time t11 and the zero lift learning execution condition is satisfied at the time t12 of the subsequent injection update timing. In this case, at time t12, the injection timing is set to a predetermined time for zero lift learning, and the injection time is set to a predetermined time for zero lift learning. Also, the zero lift learning counter is counted up by one.

The zero lift learning counter is counted up at each injection update timing, and when the zero lift learning counter exceeds the first determination value TH1 (“2” in FIG. 12), the fuel injection valve 30 during fuel cut is driven at that time t14. Is permitted, and the injector drive current is set to the predetermined current value Aoff for zero lift learning. As a result, each fuel injection valve 30 is energized with a predetermined current value Aoff within a range in which the valve body 34 is not driven at a predetermined time for zero lift learning. In FIG. 12, after the zero lift learning execution condition is satisfied, first, the fuel injection valve 30 of the second cylinder # 2 is energized for zero lift learning in the intake stroke of the second cylinder # 2, and then the fuel injection valve 30 is energized. The fuel injection valve 30 of the first cylinder # 1 is energized for zero lift learning in the intake stroke of the first cylinder # 1. Further, when the zero lift learning counter exceeds the second determination value TH2 (“4” in FIG. 12), the acquisition of the zero lift learning value is permitted at the time t16 of the subsequent injection update timing, and the electrical variation value Q1 becomes. It is calculated.

Next, a case of switching from the zero lift learning mode to the normal mode will be described with reference to FIG. In FIG. 13, when the fuel cut is completed at the time t21, the injection timing and the injection time are set to the values of the normal mode because it is determined that the zero lift learning execution condition is not satisfied at the time t22 of the injection update timing immediately after that. Can be switched. In addition, the zero lift learning counter is reset to zero, and the injector drive current is switched to the value in the normal mode.

According to the present embodiment described in detail above, the following excellent effects can be obtained.

The threshold arrival time Tdiff_1 calculated based on the negative terminal voltage Vm after the partial lift injection is turned off is the injection amount variation (mechanical variation) caused by the lift behavior of the fuel injection valve 30 and the injection caused by the circuit element of the ECU 40. Includes quantitative variability (electrical variability). On the other hand, the threshold value arrival time Tdiff_2 calculated based on the negative terminal voltage Vm after the energization pulse in the range in which the valve body 34 is not driven is turned off is information acquired when the valve body 34 is not driven, and therefore has mechanical variation. Does not include, but includes electrical variations. Therefore, according to the threshold value arrival time Tdiff_1 and the threshold value arrival time Tdiff_1, it is possible to separate the deviation amount of the injection amount due to the mechanical variation and the deviation amount of the injection amount due to the electrical variation. Therefore, it is possible to accurately grasp the variation in the injection amount of each of the fuel injection valve 30 and the ECU 40, and to perform accurate injection amount correction.

Since it is assumed that the fuel is being cut during vehicle deceleration as a condition for executing the zero lift learning, it is not necessary to execute the fuel cut for the zero lift learning. In addition, since it is a condition that the vehicle is decelerating during the fuel cut implementation period, even if fuel is injected from the fuel injection valve 30 by energization for zero lift learning, the influence on drivability should be reduced. Can be done.

Since the zero lift learning is executed on condition that the fuel pressure injected from the fuel injection valve 30 is equal to or higher than the predetermined fuel pressure that can avoid unintended fuel injection due to energization for zero lift learning, the back pressure is sufficient. Learning can be performed in a situation where the valve body 34 is high and it is difficult to open the valve. This makes it possible to prevent the fuel from being actually injected by energizing the fuel injection valve 30.

(Other embodiments)
The present invention is not limited to the contents of the above embodiment, and may be implemented as follows, for example.

In the above embodiment, the threshold arrival time Tdiff_1 calculated by lift learning is corrected by the electrical variation value Q1, and the estimated injection amount R is calculated from the corrected threshold arrival time Tdiff * Q1 to calculate the mechanical variation value Q2. However, after calculating the estimated injection amount from the threshold arrival time Tdiff_1, the mechanical variation value Q2 may be calculated by correcting the estimated injection amount with the electrical variation value Q1.

-In the above embodiment, the zero lift learning is executed during the fuel cut during deceleration of the vehicle, but the operating state of the vehicle may be any operating state in which the fuel injection valve 30 is not driven. For example, zero lift learning may be performed after the ignition is turned on and before the engine is started. Further, zero lift learning may be performed during fuel cut while the vehicle is stopped.

-In the above embodiment, the lift learning is carried out in the partial lift injection, but the lift learning may be carried out in the full lift injection.

-Each component described above is conceptual and is not limited to the above embodiment. For example, the function of one component may be distributed to a plurality of components to be realized, or the function of the plurality of components may be realized by one component.

11 ... Engine (internal combustion engine), 30 ... Fuel injection valve, 34 ... Valve body, 40 ... ECU (first information acquisition unit, second information acquisition unit, injection amount correction unit), 41 ... Microcomputer (control unit), 42 … Drive IC (control unit).

Claims (3)

It is applied to a fuel injection system of an internal combustion engine (10) including a fuel injection valve (30) and a control unit (41, 42) for controlling the drive of the fuel injection valve.
The first energization to energize the fuel injection valve is carried out in a mode in which the valve body (34) of the fuel injection valve is driven, and based on the terminal voltage of the fuel injection valve after the energization of the first energization is completed, the fuel injection valve is energized. A first information acquisition unit that acquires first drive information as information regarding the valve closing timing of the fuel injection valve, and a first information acquisition unit.
The second energization for energizing the fuel injection valve is performed in a mode in which the valve body is not driven, and the second drive information is used as information regarding the valve closing timing based on the terminal voltage after the energization of the second energization is completed. The second information acquisition unit to acquire
A correction amount calculation unit that calculates the injection correction amount of the fuel injection valve based on the first drive information acquired by the first information acquisition unit and the second drive information acquired by the second information acquisition unit. Equipped with
The correction amount calculation unit is
As the injection correction amount, a first correction amount for correcting the injection amount variation caused by the control unit and a second correction amount for correcting the injection amount variation caused by the lift behavior of the fuel injection valve are used. It is to calculate
The first correction amount is calculated from the deviation amount from the reference value for the second drive information.
Based on the first drive information and the first correction amount, information after removal correction, which is information obtained by excluding the injection amount variation caused by the control unit from the first drive information, is calculated, and the removal correction thereof is performed. A fuel injection control device for an internal combustion engine that calculates the second correction amount based on the amount of deviation from the reference value for the post-information . The fuel injection control device for an internal combustion engine according to claim 1 , wherein the second information acquisition unit performs the second energization during fuel cut during vehicle deceleration. The control unit uses an injection pulse in which the lift amount of the valve body reaches the full lift position to drive the fuel injection valve to open, and an injection pulse in which the lift amount of the valve body does not reach the full lift position. Partial lift injection that drives the fuel injection valve to open is carried out.
The fuel injection control device for an internal combustion engine according to claim 1 or 2 , wherein the first information acquisition unit performs the first energization during the execution period of the partial lift injection.

Images