JP5317022B2 - 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 having a function of calculating a variation in the injection amount of a fuel injection valve during low load operation of the internal combustion engine.

  In recent years, in order to improve the air-fuel ratio controllability of an internal combustion engine, as described in, for example, Patent Document 1 (Japanese Patent Laid-Open No. 2008-128160), exhaust gases from a plurality of cylinders of the internal combustion engine are gathered. Estimate the air-fuel ratio of each cylinder using a model that associates the detection value of the air-fuel ratio sensor installed in the flowing exhaust gas collection part (the air-fuel ratio of the exhaust gas collection part) with the air-fuel ratio of each cylinder. The cylinder-to-cylinder air-fuel ratio is calculated based on the air-fuel ratio of each cylinder, and the air-fuel ratio of each cylinder (for example, fuel injection amount) is controlled for each cylinder so as to reduce the air-fuel ratio variation of each cylinder. There is something that performs control. At that time, the injection characteristics (injection amount error) of the fuel injection valve of each cylinder based on the inter-cylinder variation of the air-fuel ratio of each cylinder detected in a plurality of operation regions (for example, a high load operation region and a low load operation region) of the internal combustion engine. ) And the fuel injection valve of each cylinder is controlled based on the learned injection characteristic of the fuel injection valve of each cylinder.

JP 2008-128160 A

  By the way, as shown in FIG. 2, the fuel injection valve of the in-cylinder internal combustion engine that injects high-pressure fuel into the cylinder has a linearity (linearity) of the change characteristic of the actual injection amount with respect to the injection pulse width (injection time). However, there is a tendency to deteriorate in a region where the injection amount is small. For this reason, during low load operation where the required injection amount during idle operation or the like is low, the variation in the injection amount of the fuel injection valve (deviation of the actual injection amount with respect to the required injection amount) tends to increase, and the injection amount of the fuel injection valve If the variation becomes large, exhaust emission may deteriorate.

  However, during low-load operation of the internal combustion engine, the amount of exhaust gas decreases and the estimation accuracy of the cylinder-by-cylinder air-fuel ratio estimation based on the output of the air-fuel ratio sensor decreases. Therefore, the air-fuel ratio variation of each cylinder cannot be obtained with high accuracy. For this reason, during the low load operation of the internal combustion engine, it is difficult to accurately correct the injection amount variation of the fuel injection valve of each cylinder, and it is difficult to reduce the injection amount variation of the fuel injection valve of each cylinder.

  Therefore, the problem to be solved by the present invention is to provide a fuel injection control device for an internal combustion engine that can learn accurately the injection amount variation of the fuel injection valve during low load operation of the internal combustion engine.

In order to solve the above-mentioned problems, an invention according to claim 1 is directed to an air-fuel ratio sensor installed in an exhaust passage of an internal combustion engine and a required injection amount of a fuel injection valve of each cylinder of the internal combustion engine (per cycle of each cylinder). Split injection in which fuel for the required injection amount is divided into a plurality of times during high load operation where the required injection amount is greater than during low load operation. Injection amount variation learning means for learning the injection amount variation of the fuel injection valve during low-load operation based on the air-fuel ratio variation obtained from the output, and this injection amount variation learning means is a predetermined injection of the fuel injection valve. when the injection number of split injection is determined based on the injection amount of the injection amount region when learning the injection quantity variation in the amount region, to learn the injection quantity variation at a plurality of injection amount region of the fuel injection valve A value obtained by dividing the required injection amount by the minimum injection amount that can be injected by the fuel injection valve (the fractional part is rounded down) is set as an initial value of the number of times of the divided injection, and the number of times of the divided injection is set to the initial value. The variation in the injection amount in the plurality of injection amount regions of the fuel injection valve is learned by learning the variation in the injection amount of the fuel injection valve by sequentially decreasing from the value.

In this configuration, by performing divided injection during high load operation where the required injection amount of the fuel injection valve of each cylinder increases, the required injection amount (injection pulse width) per injection of divided injection is reduced during low load operation. The injection amount variation of the fuel injection valve that has performed the divided injection can be made equivalent to the injection amount variation during the low load operation. During the execution of this divided injection, the air-fuel ratio is shifted by the amount of variation in the injection amount of the fuel injection valve during low load operation, resulting in air-fuel ratio variations. In addition, since the amount of exhaust gas is large during the high load operation in which this divided injection is performed, the air-fuel ratio detection accuracy of the air-fuel ratio sensor is increased. Therefore, split injection is executed during high-load operation, and the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor during execution of this split injection accurately reflects the injection amount variation of the fuel injection valve during low-load operation. Thus, by using this air-fuel ratio variation, it is possible to accurately learn the injection amount variation of the fuel injection valve during low load operation. Furthermore, when learning the injection amount variation in the predetermined injection amount region of the fuel injection valve, the number of divided injections is determined based on the injection amount in the injection amount region, so that the number of divided injections can be appropriately set. By setting, the injection amount per divided injection can be reliably set in the current injection amount region (the injection amount region in which learning is performed), and the injection amount variation in the current injection amount region can be reliably learned can do.
Further, in the present invention, when learning the injection amount variation in the plurality of injection amount regions of the fuel injection valve, a value obtained by dividing the required injection amount by the minimum injection amount that can be injected by the fuel injection valve (however, the fractional part is rounded down). Is set as the initial value of the number of divided injections, and the number of divided injections is sequentially decreased from the initial value so as to learn the variation in the injection amount of the fuel injection valve. Can be sequentially changed to sequentially change the injection amount region in which the learning is performed, and the variation in the injection amount in the plurality of injection amount regions of the fuel injection valve can be quickly learned.

  The present invention, for example, performs split injection only with a fuel injection valve of a predetermined selected cylinder during high load operation, and based on the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor during execution of the split injection, Although it is possible to learn the variation in the injection amount of the fuel injection valve of the selected cylinder during operation, in this case, to learn the variation in the injection amount of the fuel injection valve of each cylinder (all cylinders) during low load operation Therefore, it is necessary to sequentially change the selected cylinder and repeat the process of learning the injection amount variation of the fuel injection valve of the selected cylinder.

  Accordingly, as in claim 2, there is provided cylinder-by-cylinder air-fuel ratio variation calculating means for calculating the air-fuel ratio variation of each cylinder based on the output of the air-fuel ratio sensor, and dividing by the fuel injection valve of each cylinder during high load operation. Based on the air-fuel ratio variation of each cylinder calculated by the cylinder-by-cylinder air-fuel ratio variation calculation means during the execution of the divided injection, the variation in the injection amount of the fuel injection valve of each cylinder during low load operation is determined for each cylinder. You may make it learn. In this way, it is possible to learn the variation in the injection amount of the fuel injection valve of each cylinder (all cylinders) at the time of low load operation at a time.

Further, when the injection amount variation in the same injection amount region of the fuel injection valve is calculated a predetermined number of times as in claim 3 , the learning value of the injection amount variation in the injection amount region is determined based on the calculated values. You may make it do. In this way, it is possible to improve the learning accuracy of the injection amount variation of the fuel injection valve.

In this case, as in claim 4 , when the injection amount variation in the same injection amount region of the fuel injection valve is calculated a predetermined number of times under different operating conditions, the injection amount variation in the injection amount region is calculated based on those calculated values. The learning value may be determined. In this way, it is possible to provide robustness against changes in operating conditions.

FIG. 1 is a diagram showing a schematic configuration of an engine control system in Embodiment 1 of the present invention. FIG. 2 is a diagram showing the injection characteristics (relationship between the injection pulse width and the actual injection amount) of the fuel injection valve. FIG. 3 is a diagram illustrating a method for determining the number of times of divided injection. FIG. 4 is a flowchart for explaining the flow of processing of the injection amount variation learning correction routine of the first embodiment. FIG. 5 is a flowchart for explaining the flow of processing of the injection amount variation learning correction routine of the second embodiment. FIG. 6 is a flowchart for explaining the flow of processing of the injection amount variation learning correction routine of the third embodiment.

  Hereinafter, some embodiments embodying the mode for carrying out the present invention will be described.

A first embodiment of the present invention will be described with reference to FIGS.
First, a schematic configuration of the entire engine control system will be described with reference to FIG.
An air cleaner 13 is provided at the most upstream portion of the intake pipe 12 of the direct injection engine 11 that is an in-cylinder internal combustion engine, and an air flow meter 14 that detects the intake air amount is provided downstream of the air cleaner 13. Is provided. On the downstream side of the air flow meter 14, a compressor 28 of an exhaust turbine supercharger 26 described later and an intercooler 32 for cooling the intake air pressurized by the compressor 28 are provided. A supercharging pressure sensor 33 for detecting the upstream pressure (supercharging pressure) of the throttle valve 16 is provided on the downstream side of the intercooler 32, and the opening degree is adjusted by the motor 15 on the downstream side of the supercharging pressure sensor 33. There is provided a throttle valve 16 and a throttle opening sensor 17 for detecting the opening (throttle opening) of the throttle valve 16.

  Further, a surge tank 18 is provided on the downstream side of the throttle valve 16, and an intake pressure sensor 19 that detects a downstream pressure (intake pressure) of the throttle valve 16 is provided in the surge tank 18. The surge tank 18 is provided with an intake manifold 20 that introduces air into each cylinder of the engine 11, and each cylinder of the engine 11 is provided with a fuel injection valve 21 that directly injects fuel into the cylinder. Yes. An ignition plug 22 is attached to the cylinder head of the engine 11 for each cylinder, and the air-fuel mixture in the cylinder is ignited by spark discharge of the ignition plug 22 of each cylinder.

  On the other hand, in the exhaust pipe 23 (exhaust passage) of the engine 11, an air-fuel ratio sensor 24 for detecting the air-fuel ratio of the exhaust gas is provided downstream of an exhaust turbine 27 of an exhaust turbine supercharger 26 described later. A catalyst 25 such as a three-way catalyst for purifying exhaust gas is provided on the downstream side of the air-fuel ratio sensor 24.

  An exhaust turbine supercharger 26 is mounted on the engine 11. In the exhaust turbine supercharger 26, an exhaust turbine 27 is disposed upstream of the air-fuel ratio sensor 24 in the exhaust pipe 23, and a compressor is interposed between the air flow meter 14 and the throttle valve 16 in the intake pipe 12. 28 is arranged. In the supercharger 26, an exhaust turbine 27 and a compressor 28 are connected, and the exhaust turbine 27 is rotationally driven by the kinetic energy of the exhaust gas, whereby the compressor 28 is rotationally driven to supercharge intake air. .

  Further, the intake pipe 12 is provided with an intake bypass passage 29 that bypasses the upstream side and the downstream side of the compressor 28 on the upstream side of the throttle valve 16. The intake bypass passage 29 is opened and closed in the middle of the intake bypass passage 29. An air bypass valve (hereinafter referred to as “ABV”) 30 is provided. The ABV 30 is configured such that the opening / closing operation of the ABV 30 is controlled by controlling the ABV vacuum switching valve 31.

  On the other hand, the exhaust pipe 23 is provided with an exhaust bypass passage 34 that bypasses the upstream side and the downstream side of the exhaust turbine 27, and a waste gate valve (hereinafter referred to as “open”) that opens and closes the exhaust bypass passage 34 in the middle of the exhaust bypass passage 34. 35) (denoted as “WGV”). The WGV 35 is configured such that the opening degree of the WGV 35 is controlled by controlling the WGV vacuum switching valve 36 and the diaphragm type actuator 37.

  A cooling water temperature sensor 38 that detects the cooling water temperature and a knock sensor 39 that detects knocking are attached to the cylinder block of the engine 11. A crank angle sensor 41 that outputs a pulse signal every time the crankshaft 40 rotates by a predetermined crank angle is attached to the outer peripheral side of the crankshaft 40, and the crank angle and the engine are determined based on the output signal of the crank angle sensor 41. The rotation speed is detected.

  Outputs of these various sensors are input to an electronic control circuit (hereinafter referred to as “ECU”) 42. The ECU 42 is mainly composed of a microcomputer, and executes a program for engine control stored in a built-in ROM (storage medium), so that the fuel injection amount of the fuel injection valve 21 according to the engine operating state and the like. The ignition timing of the spark plug 22 is controlled.

  As shown in FIG. 2, the fuel injection valve 21 of the in-cylinder injection engine 11 that injects high-pressure fuel into the cylinder has linearity (linearity) of the change characteristic of the actual injection amount with respect to the injection pulse width (injection time). ) Tends to deteriorate in a region where the injection amount is small. For this reason, during the low load operation in which the required injection amount is reduced during idle operation or the like, the injection amount variation (deviation of the actual injection amount with respect to the required injection amount) of the fuel injection valve 21 tends to increase. When the variation in the injection amount becomes large, the exhaust emission may be deteriorated.

  Therefore, in the first embodiment, the ECU 42 executes an injection amount variation learning correction routine shown in FIG. 4 to be described later, whereby the required injection amount Qtotal of the fuel injection valve 21 of each cylinder of the engine 11 (per cycle of each cylinder). The divided injection is executed in which the fuel for the required injection amount Qtotal is divided into a plurality of times and injected during a high load operation in which the required injection amount) is larger than that during low load operation (for example, during idle operation). Based on the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor 24, the variation in the injection amount of the fuel injection valve 21 during low load operation is learned.

  By performing divided injection during high load operation where the required injection amount Qtotal of the fuel injection valve 21 of each cylinder increases, the required injection amount (injection pulse width) per injection of divided injection is reduced to required injection during low load operation. The injection amount variation of the fuel injection valve 21 that has executed the divided injection can be equivalent to the injection amount variation during the low load operation. During the execution of this divided injection, the air-fuel ratio varies by the amount of variation in the injection amount of the fuel injection valve 21 during low-load operation, resulting in variations in the air-fuel ratio. Further, since the amount of exhaust gas is large during the high load operation in which this divided injection is executed, the air-fuel ratio detection accuracy of the air-fuel ratio sensor 24 is increased. Therefore, the divided injection is executed during the high load operation, and the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor 24 during the execution of the divided injection accurately reflects the injection amount variation of the fuel injection valve 21 during the low load operation. By using this air-fuel ratio variation, it is possible to learn accurately the injection amount variation of the fuel injection valve 21 during low-load operation.

At that time, when learning the injection amount variation in the predetermined injection amount region of the fuel injection valve 21, the injection number N of the divided injections (N is an integer of 2 or more) is determined based on the injection amount Qapd in the injection amount region. To do. Specifically, as shown in FIG. 3, during the high load operation of the required injection amount Qtotal, the injection amount variation in the injection amount region including the injection amount Qapd (for example, the minimum injection amount that can be injected by the fuel injection valve 21) is changed. In the case of learning, the number N of divided injections is obtained by the following equation using the required injection amount Qtotal and the injection amount Qapd in the injection amount region in which learning is performed.
Number of injections N = Qtotal / Qapd (however, the fractional part is rounded down)
For example, when the number of injections N = Qtotal / Qapd = 3.2, the number of injections N = 3.

As a result, the number N of divided injections is set appropriately, and the injection amount (= Qtotal / N) per divided injection is set in the current injection amount region (the injection amount region in which learning is performed).
Hereinafter, the processing content of the injection amount variation learning correction routine of FIG. 4 executed by the ECU 42 will be described.

  The injection amount variation learning correction routine shown in FIG. 4 is repeatedly executed at a predetermined cycle while the ECU 42 is turned on, and serves as an injection amount variation learning means in the claims. When this routine is started, first, in step 101, it is determined whether or not the engine operating state (for example, engine speed, engine load, etc.) is in a steady state. Proceeding to step 102, the required injection amount Qtotal of the fuel injection valve 21 of each cylinder (required injection amount per cycle of each cylinder) is in a high load operation in which it is higher than in a low load operation (for example, during an idle operation). Whether or not any of the intake air amount, the intake pressure, the throttle opening, the exhaust gas amount, the required injection amount, etc. is equal to or greater than a predetermined value.

If it is determined in step 102 that the engine is operating at a high load, the routine proceeds to step 103, where the requested injection amount Qtotal and the injection amount Qapd (for example, the fuel injection valve 21 is within the injection amount region in which learning is performed). The number N of divided injections is obtained by the following equation using the minimum injection amount that can be injected).
Number of injections N = Qtotal / Qapd (however, the fractional part is rounded down)

  Note that the injection amount region in which learning is performed may be changed at every predetermined timing (for example, every time learning of the injection amount variation is completed, every predetermined time elapses, every time the system is activated, etc.).

  Thereafter, the routine proceeds to step 104, where the fuel injection valve 21 of each cylinder (all cylinders) executes N-time divided injection in which fuel for the required injection amount Qtotal is divided into N times and injected. As a result, the required injection amount (injection pulse width) per injection of the divided injection is set to the same level as the required injection amount (injection pulse width) at the time of low load operation, and the injection amount variation of the fuel injection valve 21 of each cylinder is varied. It is equivalent to the variation in injection amount during low-load operation.

  Thereafter, the process proceeds to step 105, and the output of the air-fuel ratio sensor 24 is calculated using a model in which the detected value of the air-fuel ratio sensor 24 (the air-fuel ratio of the exhaust gas flowing through the exhaust collecting portion 32) is associated with the air-fuel ratio of each cylinder. Based on this, the air-fuel ratio of each cylinder is estimated for each cylinder, and the deviation between the estimated air-fuel ratio of each cylinder and the reference air-fuel ratio (the average value or control target value of the estimated air-fuel ratio of all cylinders) is calculated. The air-fuel ratio variation is calculated for each cylinder. The processing in step 105 serves as a cylinder-by-cylinder air-fuel ratio variation calculating means in the claims. Note that the method of calculating the air-fuel ratio variation may be changed as appropriate.

  Thereafter, the routine proceeds to step 106, where the variation in the injection amount of the fuel injection valve 21 of each cylinder during the low load operation is calculated for each cylinder based on the variation in the air-fuel ratio of each cylinder. In this case, for example, when the air-fuel ratio varies X% in the rich direction, the injection amount variation is obtained as (+ X%), and when the air-fuel ratio varies Y% in the lean direction, the injection amount variation is calculated. Calculated as (−Y%). The method for calculating the injection amount variation may be changed as appropriate. The calculated value of the injection amount variation of the fuel injection valve 21 of each cylinder is used as the learning value of the injection amount variation of the fuel injection valve 21 of each cylinder in the current injection amount region, as a backup RAM (not shown) of the ECU 42, etc. Are stored in a rewritable nonvolatile memory (a rewritable memory that holds stored data even when the ECU 42 is powered off).

  On the other hand, if it is determined in step 101 that it is not in a steady state, or if it is determined in step 102 that high load operation is not being performed (that is, low load operation is being performed), the process proceeds to step 107. From the learning value data of the injection amount variation of the fuel injection valve 21 of each cylinder stored in a rewritable nonvolatile memory such as a backup RAM of the ECU 42, each cylinder in the injection amount region corresponding to the current required injection amount is stored. The learning value of the injection amount variation of the fuel injection valve 21 is read, and the injection amount of the fuel injection valve 21 of each cylinder is corrected for each cylinder using the learning value of the injection amount variation of the fuel injection valve 21 of each cylinder. In this case, for example, the basic injection amount of each cylinder may be corrected, or the final required injection amount (or injection pulse width) of each cylinder may be corrected.

  In the first embodiment described above, during the high load operation in which the required injection amount of the fuel injection valve 21 of each cylinder of the engine 11 increases, the fuel injection valve 21 of each cylinder divides the fuel for the required injection amount into a plurality of times. By executing the divided injection to inject, the required injection amount (injection pulse width) per injection of the divided injection can be made comparable to the required injection amount (injection pulse width) at the time of low load operation, The variation in the injection amount of the fuel injection valve 21 of each cylinder can be equivalent to the variation in the injection amount during low-load operation. During the execution of the split injection, the variation in the injection amount of the fuel injection valve 21 during the low load operation is calculated using the variation in the air / fuel ratio obtained from the output of the air / fuel ratio sensor 24. It is possible to accurately learn the injection amount variation of the fuel injection valve 21 at the time. Further, when the injection amount variation in the predetermined injection amount region of the fuel injection valve 21 is learned, the number N of divided injections is determined based on the injection amount Qapd in the injection amount region. The number of injections N can be set appropriately, and the injection amount per divided injection can be reliably set in the current injection amount region (the injection amount region in which learning is performed). It is possible to reliably learn the injection amount variation.

  Further, in the first embodiment, split injection is executed by the fuel injection valves 21 of each cylinder (all cylinders), and variation in the air-fuel ratio of each cylinder is performed based on the output of the air-fuel ratio sensor 24 during execution of this split injection. Since it is calculated for each cylinder and the variation in the injection amount of the fuel injection valve 21 of each cylinder during low load operation is learned for each cylinder based on the air-fuel ratio variation of each cylinder, each cylinder during low load operation ( It is possible to learn the injection amount variation of the fuel injection valves 21 of all cylinders at once.

  Further, in the first embodiment, during the low load operation of the engine 11, the injection amount of the fuel injection valve 21 of each cylinder is corrected for each cylinder using the learning value of the injection amount variation of the fuel injection valve 21 of each cylinder. Therefore, during the low load operation of the engine 11, the variation in the injection amount of the fuel injection valve 21 in each cylinder can be accurately corrected, and the variation in the injection amount of the fuel injection valve 21 in each cylinder can be sufficiently reduced.

  Next, Embodiment 2 of the present invention will be described with reference to FIG. However, description of substantially the same parts as those in the first embodiment will be omitted or simplified, and different parts from the first embodiment will be mainly described.

  In the second embodiment, the ECU 42 executes an injection amount variation learning correction routine of FIG. 5 to be described later, thereby learning the variation in the injection amount of the fuel injection valve 21 by sequentially changing the number N of divided injections. The variation in the injection amount in the plurality of injection amount regions of the fuel injection valve 21 is learned.

  In the injection amount variation learning correction routine shown in FIG. 5, if it is determined in step 201 that the engine is in a steady state, the process proceeds to step 202, where it is determined whether or not a high load operation is being performed. If it is determined, the process proceeds to step 203, and it is determined whether or not the initial value K of the number N of divided injections has been calculated during the current high load operation.

  If it is determined in step 203 that the initial value K of the number N of divided injections has not yet been calculated, the process proceeds to step 204, where the requested injection amount Qtotal and the injection amount region in which learning is performed first are performed. The initial value K of the number N of divided injections is obtained by the following equation using the injection amount Qapd (for example, the minimum injection amount that can be injected by the fuel injection valve 21).

Initial value of number of injections N K = Qtotal / Qapd (however, the decimal part is rounded down)
Thereafter, the process proceeds to step 205, where the injection number N of the current divided injection is set to an initial value K (N = K). Thereafter, the process proceeds to step 207, where it is determined whether or not the N-part divided injection in which the fuel corresponding to the required injection quantity Qtotal is divided into N parts can be executed, for example, the required injection quantity per divided injection. The determination is made based on whether (= Qtotal / N) is equal to or larger than the minimum injection amount of the fuel injection valve 21.

  If it is determined in step 207 that the N-part divided injection can be executed, the process proceeds to step 208, where the fuel injection valve 21 of each cylinder (all cylinders) divides the fuel for the required injection amount Qtotal into N times. After executing N times of divided injections, the process proceeds to step 209, where the air-fuel ratio variation of each cylinder is calculated for each cylinder based on the output of the air-fuel ratio sensor 24.

  Thereafter, the routine proceeds to step 210, where the variation in the injection amount of the fuel injection valve 21 in each cylinder during low load operation is calculated for each cylinder based on the variation in the air-fuel ratio of each cylinder, and the fuel injection valve 21 in each cylinder is calculated. The calculated value of the injection amount variation is stored in a rewritable nonvolatile memory such as a backup RAM of the ECU 42 as a learning value of the injection amount variation of the fuel injection valve 21 of each cylinder in the current injection amount region.

  Thereafter, if it is determined in step 203 that the initial value K of the number N of divided injections has already been calculated, the process proceeds to step 206, where the number N of injections of the current divided injection is reduced by one time from the previous time. (N = N-1). As a result, the required injection amount (= Qtotal / N) per injection of the divided injection is increased from the previous time, and the injection amount region in which learning is performed is changed.

  Thereafter, if it is determined that N times of divided injections can be executed, N times of divided injection is executed by the fuel injection valve 21 of each cylinder (all cylinders), and then each cylinder is determined based on the output of the air-fuel ratio sensor 24. Is calculated for each cylinder (steps 207 to 209).

  Thereafter, based on the air-fuel ratio variation of each cylinder, the variation in the injection amount of the fuel injection valve 21 in each cylinder during low load operation is calculated for each cylinder, and the variation in the injection amount of the fuel injection valve 21 in each cylinder is calculated. The value is stored in a rewritable nonvolatile memory such as a backup RAM of the ECU 42 as a learning value of the injection amount variation of the fuel injection valve 21 of each cylinder in the current injection amount region (step 210).

  On the other hand, if it is determined in step 201 that the vehicle is not in a steady state, or if it is determined in step 202 that the vehicle is not in a high load operation (that is, in a low load operation), the process proceeds to step 211. From the learning value data of the injection amount variation of the fuel injection valve 21 of each cylinder stored in a rewritable nonvolatile memory such as a backup RAM of the ECU 42, each cylinder in the injection amount region corresponding to the current required injection amount is stored. The learning value of the injection amount variation of the fuel injection valve 21 is read, and the injection amount of the fuel injection valve 21 of each cylinder is corrected for each cylinder using the learning value of the injection amount variation of the fuel injection valve 21 of each cylinder.

  In the second embodiment described above, the injection amount variation in the plurality of injection amount regions of the fuel injection valve 21 is learned by sequentially changing the injection number N of the divided injections and learning the injection amount variation of the fuel injection valve 21. Since the injection amount per one injection of the divided injection is sequentially changed, the injection amount region for performing the learning can be sequentially changed, and the injection amount variation in the plurality of injection amount regions of the fuel injection valve 21 can be changed. Can learn quickly.

  Next, Embodiment 3 of the present invention will be described with reference to FIG. However, description of substantially the same parts as those of the second embodiment will be omitted or simplified, and different parts from the second embodiment will be mainly described.

  In the third embodiment, when the ECU 42 calculates the injection amount variation in the same injection amount region of the fuel injection valve 21 by a predetermined number of times under different operating conditions by executing an injection amount variation learning correction routine of FIG. The learning value of the injection amount variation in the injection amount region is determined based on the calculated values. The routine of FIG. 6 is obtained by changing the processing of step 210 of the routine of FIG. 5 described in the second embodiment to the processing of steps 210a to 210d, and the processing of other steps is the same as that of FIG. .

  In the injection amount variation learning correction routine shown in FIG. 6, after calculating the air-fuel ratio variation of each cylinder on the basis of the output of the air-fuel ratio sensor 24 in step 209, the process proceeds to step 210a, where the air-fuel ratio variation of each cylinder is calculated. Based on this, the variation in the injection amount of the fuel injection valve 21 of each cylinder during low load operation is calculated for each cylinder.

  Thereafter, the process proceeds to step 210b, and the number of calculation of the injection amount variation in the current injection amount region is counted up. However, if the injection amount variation has been calculated up to the previous time under the same operating conditions as this time (for example, intake air amount, intake pressure, throttle opening, exhaust gas amount, required injection amount, etc.) Do not count up.

  Thereafter, the process proceeds to step 210c, and whether or not the injection amount variation in the current injection region has been calculated a predetermined number of times under different operating conditions depending on whether or not the number of times of calculation of the injection amount variation in the current injection amount region has reached a predetermined number. Determine whether.

  When it is determined in step 210c that the injection amount variation in the current injection region is calculated a predetermined number of times under different operating conditions, the process proceeds to step 210d, and the injection amount variation of the fuel injection valve 21 of each cylinder in the current injection amount region. The average value of the calculated values of the predetermined number of times is calculated for each cylinder, and the average value of the injection amount variation of the fuel injection valve 21 of each cylinder is calculated as the injection amount of the fuel injection valve 21 of each cylinder in the current injection amount region. The variation learning value is stored in a rewritable nonvolatile memory such as a backup RAM of the ECU 42.

  In the third embodiment described above, when the injection amount variation in the same injection amount region of the fuel injection valve 21 is calculated a predetermined number of times under different operating conditions, the average value of those calculated values is the injection amount in the injection amount region. Since the variation learning value is determined, it is possible to improve the learning accuracy of the injection amount variation of the fuel injection valve 21 and to provide robustness against changes in the operating conditions.

  In the third embodiment, when the injection amount variation in the same injection amount region is calculated a predetermined number of times under different operating conditions, an average value of the calculated values is determined as a learning value for the injection amount variation in the injection amount region. However, the present invention is not limited to this. For example, when the injection amount variation in the same injection amount region of the fuel injection valve 21 is calculated a predetermined number of times regardless of different operating conditions, those calculations are performed. The average value may be determined as a learning value for the injection amount variation in the injection amount region.

  In the first to third embodiments, the split injection is executed by the fuel injection valve 21 of each cylinder (all cylinders), and the empty of each cylinder is determined based on the output of the air-fuel ratio sensor 24 during the execution of the split injection. The variation in the fuel ratio is calculated for each cylinder, and the variation in the injection amount of the fuel injection valve 21 in each cylinder during low load operation is learned for each cylinder based on the variation in the air-fuel ratio of each cylinder. However, the present invention is not limited to this. For example, the split injection is executed only by the fuel injection valve 21 of the selected cylinder, and during the execution of the split injection, based on the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor 24, the selected cylinder at the time of low load operation is selected. You may make it learn the injection amount dispersion | variation of the fuel injection valve 21. FIG. In this case, the variation in the injection amount of the fuel injection valve in each cylinder (all cylinders) can be learned by sequentially changing the selected cylinder and repeating the process of learning the variation in the injection amount of the fuel injection valve in the selected cylinder. it can.

  In addition, the present invention is not limited to the in-cylinder injection type engine as shown in FIG. 1, but can be implemented with various modifications without departing from the gist, such as being applicable to an intake port injection type engine.

  DESCRIPTION OF SYMBOLS 11 ... Engine (internal combustion engine), 12 ... Intake pipe, 16 ... Throttle valve, 21 ... Fuel injection valve, 22 ... Spark plug, 23 ... Exhaust pipe (exhaust passage), 24 ... Air-fuel ratio sensor, 25 ... Catalyst, 42 ... ECU (injection amount variation learning means, cylinder-by-cylinder air-fuel ratio variation calculating means)

Claims (4)

An air-fuel ratio sensor installed in the exhaust passage of the internal combustion engine;
A split injection is performed in which the required injection amount of the fuel injection valve of each cylinder of the internal combustion engine is larger than that during the low load operation, and the divided injection is performed in which the fuel corresponding to the required injection amount is divided into a plurality of times and injected. Injection amount variation learning means for learning the injection amount variation of the fuel injection valve during the low load operation based on the air-fuel ratio variation obtained from the output of the air-fuel ratio sensor during the execution of
The injection quantity variation learning means determines the injection time number of the split injection based on the injection amount of the injection amount region when learning the injection quantity variation at a predetermined injection amount region of the fuel injection valve, the fuel When learning the injection amount variation in a plurality of injection amount regions of the injection valve, the value obtained by dividing the required injection amount by the minimum injection amount that can be injected by the fuel injection valve (however, the fractional part is rounded down) is injected in the divided injection. By setting the initial value of the number of times and learning the injection amount variation of the fuel injection valve by sequentially reducing the number of injections of the divided injection from the initial value, the injection amount variation in a plurality of injection amount regions of the fuel injection valve A fuel injection control device for an internal combustion engine characterized by learning the following . A cylinder-by-cylinder air-fuel ratio variation calculating means for calculating the air-fuel ratio variation of each cylinder based on the output of the air-fuel ratio sensor;
The injection amount variation learning means executes the divided injection by the fuel injection valve of each cylinder during the high load operation, and the air-fuel ratio of each cylinder calculated by the cylinder-by-cylinder air-fuel ratio variation calculating means during the execution of the divided injection. 2. The fuel injection control device for an internal combustion engine according to claim 1, wherein the variation in the injection amount of the fuel injection valve of each cylinder during the low load operation is learned for each cylinder based on the variation. The injection amount variation learning means, when calculating the injection amount variation in the same injection amount region of the fuel injection valve a predetermined number of times, determines a learning value of the injection amount variation in the injection amount region based on those calculated values. The fuel injection control device for an internal combustion engine according to claim 1 or 2 , wherein When the injection amount variation learning means calculates the injection amount variation in the same injection amount region of the fuel injection valve a predetermined number of times under different operating conditions, the injection amount variation learning means calculates the injection amount variation in the injection amount region based on those calculated values. 4. The fuel injection control device for an internal combustion engine according to claim 3 , wherein a learning value is determined.

Images