EP2112358A2

EP2112358A2 - Fuel injection amount learning control apparatus

Info

Publication number: EP2112358A2
Application number: EP09158441A
Authority: EP
Inventors: Masahiro Minami; Yoshiyasu Ito
Original assignee: Toyota Motor Corp
Current assignee: Toyota Motor Corp
Priority date: 2008-04-23
Filing date: 2009-04-22
Publication date: 2009-10-28
Anticipated expiration: 2029-04-22
Also published as: EP2112358B1; EP2112358A3; JP2009264165A; JP4840397B2

Abstract

When a learning value of a fuel injection amount is lost through the replacement of an ECU, an issuance device is connected to the ECU to send a forcible learning command signal, thereby forcibly making a shift to a fuel injection amount learning mode. After that, racing is carried out a plurality of times only through a single accelerator depression operation by an operator, and about 10 learning values are acquired every time racing is carried out once. When the learning values are acquired a plurality of times and learning is completed, racing is ended. Thus, a learning operation can be completed at an early stage after the replacement of the ECU. As a result, there arises no deviation in the fuel injection amount.

Description

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to a fuel injection amount learning control apparatus that is mounted on an internal combustion engine such as a diesel engine or the like to perform a learning operation for appropriately obtaining a fuel injection amount from a fuel injection valve. In particular, the invention relates to an improvement for completing a learning operation efficiently

2. Description of the Related Art

Conventionally, as disclosed in, for example, Japanese Patent Application Publication No. 2003-56389 ( JP-A-2003-56389 ) described below, in an internal combustion engine such as a diesel engine for an automobile or the like, pilot injection, namely, the injection of an extremely small amount of fuel into cylinders is carried out prior to main injection for the purpose of reducing the level of combustion noise or reducing the discharge amount of NOx.
In an engine in which this pilot injection is carried out, the optimal value of the amount of that pilot injection differs depending on the operation state of the engine at that time. In general, this pilot injection amount depends on the cylinder capacity of the engine as well, but is substantially several cubic millimeters, and pilot injection is carried out in a target pilot injection amount calculated on the basis of an engine rotational speed or the like. More specifically, open/close control (the control of an open-valve time length) of fuel injection valves (which may also be referred to hereinafter as injectors) is performed in accordance with the pressure of fuel injection.
In this case, problems such as dispersion of the amount of fuel injection resulting from individual differences among fuel injection systems (individual dispersion) and a change in the amount of fuel injection over time are caused. That is, individual differences among injectors employed in the fuel injection system (dispersion of the amount of injection), individual differences among respective sensors (dispersion of the outputs of the sensors), changes in the characteristics of the injectors over time (a deterioration in the injectors) lead to a discrepancy between a target pilot injection amount calculated by a microcomputer or the like and an actual pilot injection amount, namely, an amount of actually injected fuel. When this discrepancy arises, an appropriate pilot injection amount cannot be obtained. Then, in a situation where the actual pilot injection amount is greatly deviant from the target pilot injection amount, an increase in the level of combustion noise, an increase in the discharge amount of particulate matters (PM), misfiring of the engine, and the like may be caused. Besides, as a result of this discrepancy between the fuel injection amounts, there arises a situation where a deterioration in driveability, the generation of white smoke in exhaust gas, a deterioration in fuel consumption rate, and the like are also feared.
Thus, an operation of learning a pilot injection amount as disclosed in, for example, Japanese Patent Application Publication No. 2005-36788 ( JP-A-2005-36788 ) described below is performed. In this Japanese Patent Application Publication No. 2005-36788 ( JP-A-2005-36788 ), an operation of learning a pilot injection amount in a common rail type diesel engine is disclosed. As described above, in pilot injection, an open-valve time length of the injectors is appropriately set in accordance with a fuel injection pressure, and fuel injection is carried out in a target pilot injection amount. Thus, in an engine ECU installed in a control system of the engine, a pilot injection amount setting map in which relationships between the pilot injection amount and the time length of energization (open-valve time length) of the injectors are stored for cylinders (the injectors) respectively for a plurality of levels (e.g., 4 levels) of common rails (e.g., 30 MPa, 60 MPa, 90 MPa, 120 MPa). That is, a time length of energization of the injectors corresponding to a common rail pressure is calculated in accordance with the pilot injection amount setting map such that a target pilot injection amount determined in accordance with an engine rotational speed or the like is obtained. For example, in a four-cylinder engine in which learning is carried out respectively for four levels of common rail pressures, 16 learning values (map values) are stored in the pilot injection amount setting map.
The aforementioned operation of learning the pilot injection amount is intended to appropriately correct the learning values on the aforementioned pilot injection amount setting map, and to thereby make it possible to carry out pilot injection in an appropriate pilot injection amount even in the case of individual dispersion among the forementioned fuel injection systems or a change in the amount of injection over time.
More specifically, in this learning operation, when a command for the injectors as the amount of injection is equal to or smaller than zero (e.g., when the operating degree of an accelerator is "0" while running: when a shift-up operation or the like is performed in a vehicle mounted with a manual transmission), fuel is injected toward a specific one of the cylinders (that one of the cylinders in which a piston is located in the vicinity of a top dead center) in an extremely small amount equal to a pilot injection amount (this fuel injection will be referred to hereinafter as "single injection"), and an amount of change in engine rotational speed (an amount of change in engine operation state) resulting from this single injection or the like is recognized. Data on the amount of change in engine operation state in the case where single injection is carried out exactly in a predetermined amount are compared with an amount of change in engine operation state in the case where single injection is actually carried out. The learning value in the aforementioned pilot injection amount setting map is corrected in accordance with an amount of discrepancy between those amounts. This operation is performed respectively for the common rail pressures on the aforementioned pilot injection amount setting map and respectively for the cylinders. Thus, pilot injection can be carried out in an appropriate pilot injection amount for all the cylinders regardless of the pressure in the common rail.
FIG 9 shows, from above, changes in the deterioration degree of an injector, changes in the injection amount accuracy of the injector, and changes in learning value respectively in the case where a learning value for correcting a change in the amount of fuel injection resulting from a deterioration in the injector over time. As shown in this FIG 9, the injector deteriorates as the running distance of the vehicle increases (see a waveform representing the deterioration degree of the injector), but a learning operation is performed at predetermined intervals (see learning timings on a waveform representing the injection amount accuracy of the injector), and the learning value is updated (see a waveform representing the learning value). The injection amount accuracy of the injector is thereby appropriately obtained. A guaranteed injection amount accuracy line in FIG. 9 prescribes a permissible limit value of a deterioration in the injection amount accuracy of the injector, and the learning operation is periodically performed such that the injection amount accuracy of the injector does not exceed this guaranteed injection amount accuracy line. The timing when this learning operation is performed is set to a timing when injection is not carried out, namely, when a shift to a learning mode is made as soon as the vehicle runs by a predetermined running distance since a time point corresponding to the performance of the last learning operation and the command for the injector as the amount of injection is equal to or smaller than zero in this mode (e.g., when a shift-up operation or the like is performed in a vehicle mounted with a manual transmission).
FIG. 10 is composed of timing charts showing an example of timings when this learning operation is performed. FIG. 10 shows, from above, changes in vehicle speed, changes in engine rotational speed, and changes in the amount of injection from the injector respectively. As is apparent from this FIG. 10 as well, the aforementioned single injection is carried out once (the injection amount in this single injection is extremely small, and hence does not appear on the injection amount waveform in FIG. 10) during a decrease in engine rotational speed at a timing when the opening degree of the accelerator is temporarily "0" and the amount of fuel injection is also "0" (at the same time, a clutch is also disengaged) at the time of, for example, a shift-up operation during the running of the vehicle, and the amount of change in engine operation state at that time is recognized. The aforementioned learning value is thereby acquired, and the aforementioned pilot injection amount setting map is corrected. In FIG 10, the command for the injector as the amount of injection is "0" at each of timings t1 to t6. At each of these timings, fuel injection (single injection) is carried out in an extremely small amount to perform a learning operation.
The learning value obtained through the operation of learning the pilot injection amount as described above is stored into an engine ECU. This value is updated every time a new learning value is obtained. However, when this engine ECU breaks down and is replaced at, for example, a car dealer or a repair plant, the learning value on the aforementioned pilot injection amount setting map is lost. As a result, pilot injection cannot be carried out in an appropriate pilot injection amount.
In such a situation, after the replacement of the engine ECU, until a user drives the vehicle by a predetermined distance (after having run by the predetermined distance as one of learning operation performance conditions), the opening degree of the accelerator temporarily becomes equal to "0", the condition for performing the learning operation is fulfilled, and a new learning value is acquired, the control of the amount of fuel injection reflecting individual differences among the fuel injection systems and changes in the amount of injection over time cannot be performed. That is, the amount of fuel injection cannot be corrected on the basis of the learning value, and the vehicle continues to run with the actual pilot injection amount greatly deviant from an' appropriate amount.
FIG. 11 shows changes in the deterioration degree of the injector, changes in the injection amount accuracy of the injector, and changes in learning value respectively in a case where this engine ECU is replaced at a timing tc in FIG. 11.
As shown in this FIG. 11, at a stage before the replacement of the engine ECU, the learning operation is performed at predetermined intervals, and the learning value is updated. Thus, the injection amount accuracy of the injector is appropriately obtained (see learning timings indicated by ta and tb in FIG. 11).
However, when the engine ECU is replaced at the timing tc in FIG. 11 and there is no learning value on the aforementioned pilot injection amount setting map, the learning value is "0" in a period (a period indicated by te in FIG. 11) to a subsequent timing corresponding to the performance of the learning operation (a timing td in FIG. 11), and fuel injection is carried out with the injection amount accuracy of the injector corresponding to a deterioration line of the injector, namely, with no correction made in accordance with a deterioration in the injector or the like. That is, the deterioration in the injector or the like is directly reflected on the amount of fuel injection, and the injection amount accuracy of the injector exceeds the aforementioned guaranteed injection amount accuracy line. As a result, it is impossible to obtain an appropriate fuel injection amount. In this situation, the aforementioned inconveniences such as a deterioration in driveability, the generation of white smoke in exhaust gas, and the like are incurred.
Then, when the condition for performing the learning operation is thereafter fulfilled and the learning operation is performed, the learning operation is completed at the timing td in FIG 11. However, as described above, conventionally, the learning operation is performed only once at the timing when the opening degree of the accelerator temporarily becomes equal to "0" during the running of the vehicle at the time of a shift-up operation or the like. Therefore, a very long time is required in order to complete learning for the cylinders (the injectors) respectively at the aforementioned plurality of the levels of the common rail pressures respectively (until all the 16 learning values are acquired as described above). In the meantime, the situation incurring the aforementioned inconveniences such as a deterioration in driveability, the generation of white smoke in exhaust gas, and the like may continue. More specifically, the learning operation is performed about 10 times for a combination of one common rail pressure and one cylinder, and an average of those learning values is reflected on the pilot injection amount setting map. Therefore, in fact, all the learning values on the pilot injection amount setting map are not appropriately obtained unless the learning operation is performed, for example, about 160 times.
As described above, the learning operation (single injection) is performed only once when the opening degree of the accelerator temporarily becomes equal to "0" during the running of the vehicle because of the following reason. As this learning operation, although the opening degree of the accelerator is "0", fuel injection (the aforementioned single injection) is carried out from the injector. Therefore, when fuel injection is continuously carried out a plurality of times at a timing when the opening degree of the accelerator temporarily becomes equal to "0" during normal running of the vehicle by the user as described above, the level of combustion noise increases and may give a sense of incongruity to the user. That is, in the aforementioned diesel engine, it is not assumed that the aforementioned learning value is lost through the replacement of the engine ECU or the like, and the amount of fuel injection is learned on the basis of the idea of preventing the user from feeling a sense of incongruity as a result of an increase in the aforementioned level of combustion noise on the assumption that the amount of fuel injection does not greatly deviate from an appropriate value through the use of a previously acquired learning value even when learning is not often carried out.
Although the learning of the pilot injection amount in the diesel engine has been described above, a similar problem may also arise in the operation of learning a main injection amount or the operation of learning an amount of fuel injection in a gasoline engine.

SUMMARY OF THE INVENTION

The invention provides a fuel injection amount learning control apparatus capable of quickly acquiring a latest learning value even when a learning value is lost through the replacement of an engine ECU or the like.
In the invention, when a learning value is lost through the replacement of an engine ECU or the like, a shift to a mode in which the operation of learning the amount of fuel injection is performed is forcibly made, and the learning operation is ended in a short period of time. Thus, fuel injection is carried out at an early stage in a fuel injection amount conforming to an appropriate learning value.
An aspect of the invention relates to a fuel injection amount learning control apparatus that carries out fuel injection from a fuel injection valve into a specific cylinder of a vehicular internal combustion engine when a performance condition for a fuel injection amount learning operation is fulfilled, and performs the fuel injection amount learning operation to acquire a learning value that is based on a deviation of an actual fuel injection amount from a target fuel injection amount, based on a change in an operation state of the internal combustion engine resulting from the fuel injection. This fuel injection amount learning control apparatus is equipped with performance means for forcibly performing the fuel injection amount learning operation to acquire the learning value even when the performance condition is not fulfilled after the learning value is lost.
The fuel injection amount learning operation is forcibly performed to acquire the learning value, for example, after an ECU mounted on a vehicle is replaced with a new one. Thus, a learning value equivalent to the learning value stored in the ECU that has not been replaced yet can be acquired in a short period of time and stored into the new ECU. More specifically, a shift to a learning operation performance mode can be forcibly made at a car dealer or the like and an appropriate learning value can be obtained in a short period of time through the performance of this learning operation, without requiring the user to drive the vehicle by a relatively long distance after the replacement of the ECU (without having to await the fulfillment of the performance condition for the learning operation by causing the vehicle to run by a long distance). Thus, when the vehicle is delivered to the user, an appropriate fuel injection amount can be obtained.
As the fuel injection amount learning operation forcibly performed by the performance means, more specifically, it is possible to mention the following.
First of all, the performance means receives a forcible learning command signal from issuance means (an issuance device) for issuing a command to forcibly perform the learning operation.
In this case, the fuel injection amount learning operation is forcibly performed when the command for an injector as the amount of injection is equal to or smaller then zero during the running of the vehicle after the forcible learning command signal is received from the issuance means.
Thus, the learning operation is started substantially simultaneously with the start of the running of the vehicle. Therefore, a learning value can be acquired within a short period of time since the start of the running of the vehicle, without the necessity to cause the vehicle to run by a long distance to await the fulfillment of the performance condition for the learning operation. Accordingly, the learning operation can be completed through the driving of the vehicle by an operator of the dealer. When the learning operation is performed through the driving of the vehicle by the operator of the dealer as described above, a driving operation different from that of a general user, namely, a driving operation suited for early acquisition of the learning value can be expected. For example, in a vehicle mounted with a manual transmission, the aforementioned operator performs such a driving operation as lengthens the period in which the opening degree of the accelerator is "0" (an operation of slowly performing a shift change operation) during a shift-up operation, and a large number of learning values can thereby be acquired in a short period of time (a large number of learning values can be acquired by carrying out single injection many times in the single shift-up operation). As a result, the learning operation can be completed in a short period of time. Conventionally, the learning operation is performed during the driving of the vehicle by the general user. Therefore, a driving operation suited for early acquisition of the learning value cannot be expected, and the learning operation can be performed at most once in a single shift-up operation.
Further, it is also mentionable to forcibly perform the fuel injection amount learning operation in accordance with a learning operation start operation by an operator in a vehicle stop state after the forcible learning command signal from the issuance means is received.
In this case, the learning operation start operation performed by the operator may be an accelerator depression operation. When the accelerator depression operation is performed once by the operator, fuel injection from the fuel injection valve is automatically carried out intermittently a plurality of times to forcibly perform the fuel injection amount learning operation. Further, the fuel injection amount learning operation may be forcibly performed by carrying out racing once every time the accelerator depression operation is performed once by the operator.
In the former case of the fuel injection amount learning operation, the operation performed before the completion of the learning operation can be completely automated, and the burden on the operator can be reduced. Further, the time required to complete the learning operation can be made constant as well.
On the other hand, in the latter case of the fuel injection amount learning operation, the accelerator depression operation is repeated by the operator. In accordance with the repetition of the accelerator depression operation, the fuel injection amount learning operation is performed by carrying out fuel injection from the fuel injection valve a plurality of times. Thus, the operator can continue this learning operation while confirming a progress degree of the learning operation. Further, by suspending the accelerator depression operation before the completion of the learning operation, the learning operation can be suspended or interrupted if necessary.
As a more specific operation in the case where the fuel injection amount learning operation is performed by carrying out racing once every time the accelerator depression operation is performed once by the operator as described above, fuel injection from the fuel injection valve, which is carried out every time the accelerator depression operation is performed once by the operator, is carried out with the period of fuel injection and the amount of fuel injection set constant regardless of the operation period and the depression amount in the accelerator depression operation performed by the operator.
Thus, even when the operation period or the operation amount (depression amount) in the accelerator depression operation performed by the operator is dispersed, the learning operation can be performed with the period of fuel injection and the amount of fuel injection set constant (in a certain fuel injection period and in a certain fuel injection amount). That is, fuel injection suited for early acquisition of the learning value can be carried out regardless of dispersion of the accelerator depression operation performed by the operator.
Further, in the case where the fuel injection amount learning operation is forcibly performed by automatically carrying out fuel injection from the fuel injection valve intermittently in a vehicle stop state when the forcible learning command signal is received from the issuance means, the learning operation can be entirely automatically completed without requiring any special operation such as depression of the accelerator by the operator or the like. As a result, the burden on the operator can be drastically reduced,
In the invention, when the learning value is lost due to the replacement of the ECU or the like, a shift to a mode in which the fuel injection amount learning operation is performed is forcibly made. Thus, fuel injection in a fuel injection amount conforming to an appropriate learning value can be carried out at an early stage by completing the learning operation in a short period of time. As a result, the inconveniences resulting from the deviation of the fuel injection amount from an appropriate value can be avoided.

BRIEF DESCRIPTION OF THE DRAWINGS

The features, advantages, and technical and industrial significance of this invention will be described in the following detailed description of example embodiments of the invention with reference to the accompanying drawings, in which like numerals denote like elements, and wherein:

FIG 1 is a schematic constructional view of an engine according to the embodiments of the invention and a control system thereof;
FIG 2 is composed of views showing an example of changes in deterioration degree of an injector, changes in injection amount accuracy of the injector, and changes in learning value respectively in the first embodiment of the invention;
FIG 3 is a waveform chart showing changes in engine rotational speed in the case where single injection is carried out 10 times during a decrease in engine rotational speed to acquire 10 learning values;
FIG 4 is composed of views showing an example of changes in engine rotational speed and changes in injection amount of an injector in the second embodiment of the invention;
FIG. 5 is a flowchart showing a control procedure in a learning operation in the third embodiment of the invention;
FIG. 6 is composed of views showing an example of changes in progress degree of learning, changes in engine rotational speed, changes in injection amount of an injector, and changes in opening degree of an accelerator respectively in the third embodiment of the invention;
FIG 7 is composed of views showing an example of changes in progress degree of learning, changes in engine rotational speed, changes in injection amount of an injector, and changes in opening degree of an accelerator respectively in a comparative example;
FIG 8 is a flowchart showing a control procedure in a learning operation in the fourth embodiment of the invention;
FIG 9 is composed of views showing an example of changes in deterioration degree of an injector, changes in injection amount accuracy of the injector, and changes in learning value respectively in the case where learning values are acquired without replacing an ECU;
FIG 10 is composed of timing charts showing an example of timings when a conventional learning operation is performed; and
FIG 11 is composed of views showing changes in deterioration degree of an injector, changes in injection amount accuracy of the injector, and changes in learning value respectively in the case where an ECU is replaced.

DETAILED DESCRIPTION OF EMBODIMENTS

Embodiments of the invention will be described hereinafter on the basis of the drawings. In these embodiments of the invention, a case where the invention is applied to a common rail-equipped cylinder direct injection-type multi-cylinder (e.g., four-cylinder) diesel engine mounted on an automobile will be described.
- Description of Construction of Engine - First of all, the overall construction of the diesel engine (hereinafter referred to simply as the engine) according to these embodiments of the invention will be described. FIG 1 is a schematic constructional view of an engine 1 according to these embodiments of the invention and a control system thereof.
As an intake system, an intake passage 4 is connected via an intake valve 4a to a combustion chamber 3 formed between a cylinder 1a and a piston 1b in this engine 1. This intake passage 4 is provided from an upstream side thereof with an air cleaner 6 for filtering intake air, an intake air amount sensor 8 for detecting an intake air amount, an intake air temperature sensor 10 for detecting a temperature of intake air, and a throttle valve 14 for adjusting an amount of intake air introduced into the combustion chamber 3.
The throttle valve 14 is driven by a drive mechanism 16 to be opened/closed. This drive mechanism 16 is constructed with a step motor 18 and a gear group for drivingly coupling this step motor 18 to the throttle valve 14. The step motor 18 is drivingly controlled by an electronic control unit (hereinafter referred to as "the ECU") 20 for performing various types of control of the engine 1. Further, the drive mechanism 16 is provided with a fully open switch 22 that is turned on when the throttle valve 14 assumes a fully open position.
On the other hand, as an exhaust system, an exhaust passage 24 is connected to the combustion chamber 3 via an exhaust valve 24a. An exhaust gas recirculation (EGR) passage 26 branches off from this exhaust passage 24. This EGR passage 26 is connected to the intake passage 4 downstream of the throttle valve 14. The EGR passage 26 is provided with an EGR valve 30 that is driven by an actuator 28, which is controlled by the ECU 20, to be opened/closed. By adjusting the amount of intake air by the throttle valve 14 and the amount of EGR by this EGR valve 30, the ratio between the amount of intake air introduced into the combustion chamber 3 and the amount of EGR can be freely set. Thus, the intake air amount and the EGR amount can be appropriately controlled over an entire operation range of the engine 1.
The engine 1 is provided with a plurality of cylinders #1, #2, #3, and #4 (although there are four cylinders in this embodiment of the invention, only one of them is shown), and an injector 32 is disposed for the combustion chamber 3 of each of the cylinders #1 to #4. Fuel injection from the injector 32 into each of the cylinders #1 to #4 of the engine 1 is controlled by turning an injection control electromagnetic valve 32a on/off.
The injector 32 is connected to a common rail 34 as a pressure accumulation container common to the respective cylinders. While the injection control electromagnetic valve 32a is open (in an injector open-valve period), fuel in the common rail 34 is injected into the combustion chamber 3 by the injector 32. A relatively high pressure equivalent to a fuel injection pressure is accumulated in the common rail 34. In order to realize this pressure accumulation, the common rail 34 is connected to a discharge port 36a of a supply pump 36 via a supply pipe 35. Further, a check valve 37 is provided such that the supply pipe 35 extends thereacross. Due to the presence of this check valve 37, fuel is allowed to be supplied from a supply pump 36 to the common rail 34, and fuel is stopped from flowing backward from the common rail 34 to the supply pump 36.
The supply pump 36 is connected to a fuel tank 38 via an intake port 36b. A filter 39 is provided such that the supply pump 36 extends thereacross. The supply pump 36 sucks in fuel from the fuel tank 38 via the filter 39. Further, at the same time, the supply pump 36 receives a rotational driving force from a crankshaft as an output shaft of the engine 1 to cause a plunger to move in a reciprocating manner, raises the pressure of fuel to a required pressure, and supplies high-pressure fuel to the common rail 34.
In addition, a pressure control valve 40 is provided in the vicinity of the discharge port 36a of the supply pump 36, This pressure control valve 40 controls the pressure of fuel discharged from the discharge port 36a to the common rail 34 (i.e., injection pressure). When this pressure control valve 40 is opened, surplus fuel that is not discharged from the discharge port 36a is returned from a return port 36c provided through the supply pump 36 to the fuel tank 38 via a return pipe (return flow passage) 41.
As described above, a fuel supply system of the engine 1 is mainly composed of the fuel tank 38, the supply pump 36, the common rail 34, and the injector 32.
Further, a glow plug 42 is disposed in the combustion chamber 3 of the engine 1. This glow plug 42 is a start assist device that glows when a current is caused to flow to a glow relay 42a thereof immediately before starting the engine 1. When part of fuel spray is blown on the glow relay 42a, ignition and combustion are thereby promoted.
The crankshaft of the engine 1 is provided with a rotor that rotates in synchronization with rotation of this crankshaft, and with a rotational speed sensor 44 constructed as an electromagnetic pickup that detects a convex portion formed on an outer peripheral face of that rotor and outputs a pulse signal corresponding to a rotational speed thereof. An output of this rotational speed sensor 44 is fetched by the ECU 20 as a signal contributing to the calculation of the rotational speed of the engine 1.
The electronic control unit (ECU) 20 is equipped with a central processing unit (CPU), a read only memory (ROM), a random access memory (RAM), a backup RAM, a timer, a counter, and the like. The ECU 20 is constructed by connecting these components, an external input circuit including analog/digital (A/D) converters, and an external output circuit including analog/digital (A/D) converters through a bidirectional bus.
The ECU 20 constructed as described above has detection signals of various sensors input thereto via the external input circuit, and performs various types of control regarding the operation state of the engine 1, such as basic control on fuel injection in the engine 1 or the like, on the basis of those signals. More specifically, in addition to intake air amount information detected by the intake air amount sensor 8 and intake air temperature information detected by the intake air temperature sensor 10, pieces of information such as accelerator opening degree information (information on the depression amount of an accelerator pedal) detected by an accelerator opening degree sensor 46, on/off information on an ignition (IG) switch 48, on/off information on a starter switch 50, coolant temperature information detected by a coolant temperature sensor 52 provided in a water jacket 2a, shift position information detected by a shift position sensor 54 provided on a transmission, vehicle speed information detected through a signal of a vehicle speed sensor 56, fuel temperature information detected by a fuel temperature sensor 58 provided in a return pipe 41 extending from the injector 32, fuel temperature information detected by a fuel temperature sensor 59 provided in the vicinity of the return port 36c, information on the pressure of fuel (injection pressure PC) detected by a fuel pressure sensor 60 provided in the common rail 34, and the like are fetched as well by the ECU 20. The ECU 20 performs the various types of control on the operation state of the engine 1 on the basis of these pieces of information.
- Fuel Injection Amount Learning Control - Next, fuel injection amount learning control as a control operation constituting a feature of these embodiments of the invention will be described.
In fuel injection amount learning control, an extremely small amount of fuel is injected into a specific one of the cylinders (that one of the cylinders in which the piston 1b is in the vicinity of a compression top dead center) when the command for the injector 32 as the amount of injection is equal to or smaller than zero as described above, and an amount of change in engine operation state resulting from this single injection is recognized. Then, data on the amount of change in engine operation state in the case where single injection is carried out exactly in a predetermined amount are compared with an amount of change in engine operation state in the case where single injection is actually carried out. In accordance with the amount of deviation, a learning value of the fuel injection amount is acquired, and the learning value on an injection amount setting map stored in the ECU 20 is corrected.
However, when the ECU 20 is replaced at a car dealer or the like due to a failure in the ECU 20, the learning value on the aforementioned injection amount setting map is lost. As a result, it may be impossible to obtain an appropriate injection amount.
In each of the following embodiments of the invention, under such a situation, a learning operation is completed at an early stage after the replacement of the ECU 20. Thus, an appropriate learning value can be stored into the newly mounted ECU 20 in a short period of time.
In a situation in which the ECU 20 is not replaced, the learning operation is performed at a timing similar to that of the conventional art. That is, a shift to a learning mode is made when the vehicle runs by a predetermined running distance since a time point corresponding to the completion of the last learning operation. In this mode, single injection is carried out when the command for the injector 32 as the amount of injection is equal to or smaller than zero, and a learning value of the fuel injection amount is thereby acquired.
(First Embodiment) In this embodiment of the invention, after the ECU (engine ECU) 20 is replaced at a car dealer due to a failure in the ECU 20 or the like, a learning operation is performed by causing the vehicle to run through the driving thereof by an operator at the car dealer. Thus, a learning value of the amount of fuel injection is acquired.
More specifically, an issuance device (which can be regarded as the issuance means of the invention: e.g., a personal computer) that issues a command to forcibly cause this newly mounted ECU 20 to perform the learning operation (a forcible learning command signal) is connected to the newly mounted ECU 20, and a shift to a fuel injection amount learning mode is forcibly made (an operation of forcibly performing a fuel injection amount learning operation by the performance means). Then, while the operator at the car dealer causes the vehicle to run by a relatively short distance (e.g., about 2 km), a learning value of the fuel injection amount is appropriately obtained.
FIG. 2 shows an example of changes in deterioration degree of the injector, changes in injection amount accuracy of the injector 32 (the accuracy of the amount of injection deteriorates as the distance from the axis of abscissa increases in FIG. 2), and changes in learning value in this case where the ECU 20 has been replaced at a timing Tc in FIG. 2.
As shown in this FIG. 2, before the ECU 20 is replaced, the injector 32 deteriorates as the running distance of the vehicle increases (see a waveform representing the deterioration degree of the injector 32), but the learning operation is performed at predetermined intervals (see learning timings Ta and Tb on a waveform representing the injection amount accuracy of the injector 32). Since the learning value of the fuel injection amount is updated (see a waveform representing the learning value), the injection amount accuracy of the injector 32 is appropriately obtained. As the interval for performing the learning operation in this case, the learning operation is performed when the command for the injector 32 as the amount of injection is equal to or smaller than zero (e.g., when a shift-up operation is performed in a vehicle mounted with a manual transmission) after the vehicle has run by a predetermined running distance since the performance of the last learning operation. The learning operation is performed at such a timing, and as a result, the injection amount accuracy of the injector 32 does not exceed a guaranteed injection amount accuracy line as a permissible limit value of a deterioration in the injection amount accuracy of the injector 32.
Then, when the ECU 20 is replaced at the timing Tc in FIG. 2 and there is no learning value on the pilot injection amount setting map, the issuance device is connected to the newly mounted ECU 20, and a command to forcibly cause the ECU 20 to perform the learning operation is transmitted from this issuance device to the ECU 20. Thus, the ECU 20 forcibly shifts to a fuel injection amount learning mode.
In a state where the shift to the fuel injection amount learning mode has been made as described above, the learning operation is performed while the operator at the car dealer causes the vehicle to run by a relatively short distance (e.g., about 2 km).
In this learning operation, the fuel injection amount learning operation is performed substantially simultaneously with the start of the running of the vehicle. Therefore, there is no need to cause the vehicle to run by a long distance to await the fulfillment of a performance condition for the learning operation. Accordingly, the learning operation can be completed through the driving of the vehicle by a relatively short distance by the operator at the dealer.
Further, in the case where the learning operation is performed through the driving by the operator at the dealer as described above, a driving operation different from that of general user, namely, a driving operation suited for the acquisition of the learning value of the fuel injection amount at an early stage can be expected. For example, in a vehicle mounted with a manual transmission, the aforementioned operator performs such a driving operation as lengthens the period in which the opening degree of the accelerator is "0" (an operation of slowly performing a shift change operation) during a shift change (e.g., shift-up) operation. A large number of (e.g., 10) learning values can thereby be acquired in the single shift change operation. As a result, the learning operation can be completed in a short period of time. For example, the performance of such a driving operation as lengthens the period in which the opening degree of the accelerator is "0" during the performance of the learning operation is mentioned in an operation manual. The operator performs the driving operation conforming to the operation manual, and the completion of the learning operation in a short period of time can thereby be realized with ease.
FIG. 3 is a waveform chart showing changes in engine rotational speed in the case where single injection is carried out 10 times during a decrease in engine rotational speed when the opening degree of the accelerator is set equal to "0" during a shift-up operation or the like (the opening degree of the accelerator is returned to "0" at a timing Td in FIG 3). By performing such a driving operation as ensures a long period in which the opening degree of the accelerator is "0" (a period Te in FIG 3) as described above, single injection can be carried out 10 times. Thus, 10 learning values can be continuously acquired during the single shift-up operation.
More specifically, when it is determined on the basis of an accelerator opening degree signal detected by the accelerator opening degree sensor 46 that the opening degree of the accelerator has become equal to "0", an engine rotational speed at the present time point is recognized through a signal from the rotational speed sensor 44, and a number of times of single injection that can be carried out during a decrease from the rotational speed to an idle rotational speed is calculated. Then, while the engine rotational speed decreases since the opening degree of the accelerator has become equal to "0", single injection is intermittently carried out the number of times calculated through the aforementioned calculation. The learning value of the fuel injection amount is thereby continuously acquired. FIG 3 shows a case where an engine rotational speed N1 before the start of a shift-up operation is 4000 rpm, an engine rotational speed N2 (idle rotational speed) during the shift-up operation (when the opening degree of the accelerator is "0") is 800 rpm, fuel of 2 mm³ is injected when single injection is curried out once, and the engine rotational speed rises by 2 to 3 rpm (N3 in FIG. 3) through individual single injection. In this case, single injection can be carried out 10 times. When the difference between the engine rotational speed N1 before the start of the aforementioned shift-up operation and the engine rotational speed N2 during the shift-up operation is increased, single injection can be carried out more than 11 times. More than 11 learning values can also be acquired continuously while performing the shift-up operation once. For example, in a vehicle mounted with a manual transmission, it is conceivable to make an operation of disengaging a clutch precede an operation of releasing depression of an accelerator pedal during shift-up to temporarily make the rotational speed of an engine soar and hence hold the engine rotational speed high at the time point of the release of depression of the accelerator pedal.
A large number of learning values can be continuously acquired during a single shift change operation as described above. Therefore, in the case where a learning operation is completed by acquiring, for example, 160 learning values (the learning operation is performed for each of cylinders at four levels of common rail pressures (e.g., 30 MPa, 60 MPa, 90 MPa, and 120 MPa) to acquire a learning value of the amount of fuel injection), for example, 10 learning values can be acquired through a single shift change operation. As a result, the learning operation can be completed by performing the shift change operation only 16 times. Conventionally, the learning operation is performed during the driving by a general user. Therefore, a driving operation suited for early acquisition of a learning value cannot be expected, and the learning operation is performed at most once through a single shift change operation. Thus, the shift change operation needs to be performed 160 times to acquire 160 learning values, and a long period of time is required to complete the learning operation.
As described above, in this embodiment of the invention, the learning operation can be completed when the operator at the dealer causes the vehicle to run by a relatively short distance. Therefore, the latest learning value is acquired substantially simultaneously with or with a slight delay from the timing tc in FIG. 2, and high injection amount accuracy of the injector 32 can be obtained owing to this learning value. That is, even when the ECU 20 is replaced, the learning operation can be performed in the same manner as or at an earlier timing than in the case where the ECU 20 is not replaced. As a result, an appropriate fuel injection amount can be obtained.
Alternate long and two short dashes lines in FIG. 2 indicate a change in the accuracy of fuel injection in a conventional example, when the learning value is set to "0" until a subsequent condition for performing the learning operation is fulfilled due to the replacement of the ECU 20, and the accuracy of fuel injection has deteriorated.
As described above, in this embodiment of the invention, the learning operation can be completed at an early stage through the driving by the operator at the dealer. In delivering the vehicle to the user, an appropriate fuel injection amount can be ensured. As a result, inconveniences (a deterioration in driveability, the generation of white smoke in exhaust gas, and the like) resulting from a deviation of the amount of fuel injection from an appropriate value can be avoided.
(Second Embodiment) In this embodiment of the invention, after the ECU (engine ECU) 20 is replaced at a car dealer due to a failure in the ECU 20 or the like, a learning operation is performed in a stop state of the vehicle to acquire a learning value of the amount of fuel injection. That is, the learning operation is completed through so-called racing, namely, by intermittently performing a fuel injection operation to intermittently raise the rotational speed of the engine. Now, what is different from the foregoing first embodiment of the invention will be mainly described.
More specifically, after the aforementioned ECU 20 is replaced, an issuance device for issuing a command to forcibly cause this newly mounted ECU 20 to perform the learning operation is connected to the newly mounted ECU 20 to forcibly make a shift to a fuel injection amount learning mode. Then, with the engine 1 in an idling state and the transmission in a neutral state, an operator of the car dealer performs an operation of depressing the accelerator pedal (which can be regarded as the operation of starting the learning operation of the invention) only once while the vehicle remains in a stop state. The learning operation is thereby started. That is, with the transmission left in the neutral state, the ECU 20 automatically carries out racing a plurality of times to acquire the learning value of the fuel injection amount as soon as this operation of depressing the accelerator pedal is performed once.
FIG. 4 shows an example of changes in engine rotational speed and changes in the injection amount of the injector 32 in this case.
As shown in this FIG. 4, when the operator performs the operation of depressing the accelerator pedal once, racing is carried out 14 times. Every time racing is carried out, single injection is carried out 10 times during a decrease in the engine rotational speed to continuously acquire 10 learning values, as in the case described using FIG. 3 in the foregoing first embodiment of the invention. In this case, the fuel injection period and the fuel injection amount remain unchanged whenever racing is carried out, and the learning operation can be completed at an early stage.
Thus, 140 learning values can be acquired, and the learning operation can be completed substantially as soon as racing is carried out 14 times (in the case where the learning operation is completed by acquiring 140 learning values).
Further, in this embodiment of the invention, the operation before the completion of the learning operation can be automated, and the burden on the operator can be alleviated. Further, the time required to complete the learning operation can be made constant as well.
(Third Embodiment) In this embodiment of the invention as well as the foregoing second embodiment of the invention, after the ECU 20 is replaced at a car dealer due to a failure in the ECU 20 or the like, a learning operation is performed in a stop state of the vehicle to acquire a learning value of the fuel injection amount. That is, the learning operation is completed through so-called racing, namely, by intermittently raising the engine rotational speed. Now, what is different from the foregoing second embodiment of the invention will be mainly described.
More specifically, after the ECU 20 is replaced, an issuance device for issuing a command to forcibly cause this newly mounted ECU 20 to perform the learning operation is connected to the newly mounted ECU 20 to forcibly make a shift to a fuel injection amount learning mode. Then, every time an operator of the car dealer performs the operation of depressing the accelerator pedal once with the vehicle left in a stop state, the ECU 20 automatically carries out racing once to acquire learning values of the fuel injection amount. That is, every time the operator intermittently performs the operation of depressing the accelerator pedal a plurality of times, racing is carried out. Thus, a large number of learning values are acquired.
A concrete procedure in this operation will be described with reference to a flowchart of FIG. 5. First of all, it is determined in step ST1 whether or not a command signal from the issuance device (a command signal for forcibly performing the learning operation) has been input. When this command signal has been input and the result of the determination in step ST1 is Yes, a transition to step ST2 is made to make a shift to a fuel injection amount learning mode based on racing.
After the shift to the fuel injection amount learning mode is thus made, an operation of depressing the accelerator pedal by the operator is awaited in step ST3. When the operation of depressing the accelerator pedal is performed, the result of the determination in step ST3 is Yes, and a transition to step ST4 is made. In this step ST4, a racing operation is performed once. Due to the performance of this racing operation, 10 learning values are acquired as in the case of, for example, the foregoing embodiments of the invention.
After the performance of this racing operation, a transition to step ST5 is made to make a determination on the completion of the learning operation. For example, the determination on the completion of the learning operation is made depending on whether or not 140 learning values have been acquired. When the number of the acquired learning values is still small and the result of the determination in step ST5 is No, a return to step ST3 is made. That is, a subsequent operation of depressing the accelerator pedal by the operator is awaited. Then, when the operation of depressing the accelerator pedal is performed, the racing operation is performed once as in the aforementioned case to acquire, for example, 10 learning values. In this case as well, the fuel injection period and the fuel injection amount remain unchanged whenever racing is carried out, as in the case of the foregoing second embodiment of the invention.
When this operation is repeated and' a predetermined number of (e.g., 140) learning values are acquired, the result of the determination in step ST5 is Yes, and a racing learning mode is canceled. A recovery to a normal mode is then made to end the learning operation.
FIG. 6 shows an example of changes in progress degree of learning, changes in engine rotational speed, changes in the injection amount of the injector 32, and changes in the opening degree of the accelerator in this case.
As shown in this FIG. 6, in this embodiment of the invention, even when the operation period and the operation amount (depression amount) in the operation of depressing the accelerator pedal by the operator is dispersed, the learning operation is performed with the fuel injection period and the fuel injection amount held constant. That is, regardless of dispersion of the operation of depressing the accelerator pedal by the operator, fuel injection suited for early acquisition of the learning value of the fuel injection amount can be carried out.
FIG 7 shows an example in which the learning operation is performed by carrying out fuel injection with the fuel injection period and the fuel injection amount that conform to the operation period and the operation amount (depression amount) in the operation of depressing the accelerator by the operator. That is, when the operation period of the accelerator depression operation performed by the operator is short, the period of fuel injection is also short. When the operation amount of the accelerator depression operation (depression amount) is small, the amount of fuel injection is also small. In this case, the progress degree of learning may slow down. In this embodiment of the invention, the learning operation is performed with the fuel injection period and the fuel injection amount held constant regardless of dispersion of the accelerator depression operation performed by the operator as described above. Therefore, the progress degree of learning is high, and the learning operation can be completed in a short period of time.
As described above, in this embodiment of the invention, the fuel injection amount learning operation is performed by carrying out fuel injection from the injector 32 a plurality of times through the repetition of the accelerator depression operation by the operator. Therefore, the operator can continue this learning operation while recognizing the progress degree of the learning operation. Further, it is also possible to stop or suspend the learning operation in case of necessity by suspending the accelerator depression operation before the completion of the learning operation.
(Fourth Embodiment) In this embodiment of the invention as well as the foregoing second embodiment of the invention, a learning value of the fuel injection amount is acquired by performing a learning operation in a stop state of the vehicle after the replacement of the ECU 20 at a car dealer due to a failure in the ECU 20 or the like. That is, the learning operation is completed through so-called racing, namely, by intermittently raising the rotational speed of the engine. Now, what is different from the foregoing second embodiment of the invention will be mainly described.
More specifically, after the ECU 20 is replaced, an issuance device for issuing a command to forcibly cause this newly mounted ECU 20 to perform the learning operation is connected to the newly mounted ECU 20 to forcibly make a shift to a fuel injection amount learning mode. Then, the learning operation is started substantially simultaneously with the reception of a command signal from this issuance device. That is, the ECU 20 automatically carries out racing a plurality of times to acquire the learning value of the fuel injection amount.
A concrete procedure in this operation will be described with reference to a flowchart of FIG. 8. First of all, it is determined in step ST11 whether or not a command signal from the issuance device (a command signal for forcibly performing the learning operation) has been input. When this command signal has been input and the result of the determination in step ST11 is Yes, a transition to step ST12 is made to make a shift to a fuel injection amount learning mode (racing learning mode).
As soon as the shift to the fuel injection amount learning mode is thus made, racing is automatically carried out a plurality of times. Thus, the operation of learning the fuel injection amount is started, and learning values of the fuel injection amount are acquired. After this racing operation is performed, a shift to step ST13 is made to make a determination on the completion of the learning operation. For example, the determination on the completion of the learning operation is made depending on whether or not, for example, 140 learning values have been acquired. When the number of the acquired learning values is still small and the result of the determination in step ST13 is No, the learning operation is continued.
Then, when a predetermined number of learning values are acquired, the result of the determination in step ST13 is Yes, and the fuel injection amount learning mode is cancelled. Thus, a recovery to a normal mode is made to end the learning operation.
In this case as well, as shown in the aforementioned FIG. 4, when the command signal from the issuance device is received, racing is automatically carried out 14 times. Every time racing is carried out, single injection is carried out 10 times during a decrease in the engine rotational speed to continuously acquire 10 learning values, as in the case described using FIG. 3 in the foregoing first embodiment of the invention.
Thus, 140 learning values can be acquired, and the learning operation can be completed substantially as soon as racing composed of a 14-time racing operation is carried out. In one racing, single injection is carried out 10 times.
As described above, in this embodiment of the invention, the operation before the completion of the learning operation can be entirely automated, and the burden on the operator can be alleviated. Further, the time required to complete the learning operation can be made constant as well.
- Other Embodiment - In each of the embodiments of the invention described above, the case where the invention is applied to the common rail-equipped in-cylinder direct injection-type multi-cylinder (four-cylinder) diesel engine is described. The invention is not limited to this case, but is also applicable to other diesel engines having an arbitrary number of cylinders, for example, a six-cylinder diesel engine and the like. Further, the invention is also applicable to other internal combustion engines such as gasoline engines and the like as well as diesel engines. Furthermore, the engines to which the invention is applicable are not necessarily designed for automobiles.
Further, in each of the foregoing embodiments of the invention, the case where single injection is carried out 10 times during a decrease in the engine rotational speed in performing the learning operation is described. However, single injection should not necessarily be carried out 10 times.
Further, in each of the foregoing embodiments of the invention, the case where the invention is applied to the vehicle mounted with the manual transmission is described. The invention is not limited to this case, but is also applicable to a vehicle mounted with an automatic transmission. For example, when the learning operation is performed while causing the vehicle to run as in the case of the foregoing first embodiment of the invention, the learning operation can be completed at an early stage by driving the vehicle in such a manner as to lengthen the period in which the accelerator opening degree is "0". For example, a driving operation that lengthens the period of freewheel running is performed. On the other hand, when the learning operation is performed in the stop state of the vehicle as in the cases of the foregoing second to fourth embodiments of the invention, a shift lever of the transmission remains at a range position of "P (parking)" or "N (neutral)".
Moreover, in the foregoing second embodiment of the invention, when the operator performs the operation of depressing the accelerator pedal once, racing is carried out a plurality of (14) times. That is, the accelerator pedal is utilized as a switch for carrying out racing a plurality of times. Thus, some operation as well as the operation of depressing the accelerator pedal can be utilized as a switch for carrying out racing a plurality of times. For example, racing can be carried out a plurality of times in response to an operation of depressing a clutch pedal, or racing can be carried out a plurality of times when this operation of depressing the clutch pedal and the operation of depressing a brake pedal are both performed. Further, racing may be carried out a plurality of times in response to the operation of a switch group provided on the issuance device. When a learning value of a fuel injection amount is lost through the replacement of an ECU, an issuance device is connected to the ECU to send a forcible learning command signal, thereby forcibly making a shift to a fuel injection amount learning mode. After that, racing is carried out a plurality of times only through a single accelerator depression operation by an operator, and about 10 learning values are acquired every time racing is carried out once. When the learning values are acquired a plurality of times and learning is completed, racing is ended. Thus, a learning operation can be completed at an early stage after the replacement of the ECU. As a result, there arises no deviation in the fuel injection amount.

Claims

A fuel injection amount learning control apparatus that carries out fuel injection from a fuel injection valve into a specific cylinder of a vehicular internal combustion engine when a performance condition for a fuel injection amount learning operation is fulfilled, and performs the fuel injection amount learning operation to acquire a learning value that is based on a deviation of an actual fuel injection amount from a target fuel injection amount, based on a change in operation state of the internal combustion engine resulting from the fuel injection, characterized by comprising:
performance means (20) for forcibly performing the fuel injection amount learning operation to acquire the learning value even when the performance condition is not fulfilled after the learning value is lost.
The fuel injection amount learning control apparatus according to claim 1, further comprising
issuance means for issuing a command to forcibly perform the learning operation, wherein
the performance means (20) makes a shift to a learning mode and forcibly performs the fuel injection amount learning operation by receiving a forcible learning command signal from the issuance means.
The fuel injection amount learning control apparatus according to claim 2, wherein
the performance means (20) forcibly performs the fuel injection amount learning operation when the command for an injector as the amount of injection is equal to or smaller then zero during running of a vehicle, after receiving the forcible learning command signal from the issuance means.
The fuel injection amount learning control apparatus according to claim 3, wherein the vehicle is mounted with a manual transmission,
the performance means (20) makes an operation of disengaging a clutch precede an operation of releasing depression of an accelerator pedal at a time of shift-up to thereby temporarily make an engine rotational speed soar and heighten the engine rotational speed at a time point when depression of the accelerator pedal is released, and calculates how many times single injection can be carried out before the heightened engine rotational speed decreases to an idle rotational speed.
The fuel injection amount learning control apparatus according to claim 2, wherein
the performance means (20) forcibly performs the fuel injection amount learning operation in accordance with a learning operation start operation by an operator in a vehicle stop state after receiving the forcible learning command signal from the issuance means.
The fuel injection amount learning control apparatus according to claim 5, wherein
the learning operation start operation performed by the operator is an accelerator depression operation, and
the performance means forcibly performs the fuel injection amount learning operation by automatically carrying out fuel injection from the fuel injection valve intermittently a plurality of times every time the accelerator depression operation is performed once by the operator.
The fuel injection amount learning control apparatus according to claim 5, wherein
the learning operation start operation performed by the operator is an accelerator depression operation, and
the performance means (20) forcibly performs the fuel injection amount learning operation by carrying out racing once every time the accelerator depression operation is performed once by the operator.
The fuel injection amount learning control apparatus according to any one of claims 5 to 7, wherein
the performance means (20) carries out fuel injection from the fuel injection valve, which is carried out every time the accelerator depression operation is performed once by the operator, with a fuel injection period and a fuel injection amount set constant regardless of an operation period and a depression amount in the accelerator depression operation performed by the operator.
The fuel injection amount learning control apparatus according to claim 2, wherein the performance means (20) forcibly performs the fuel injection amount learning operation by automatically carrying out fuel injection from the fuel injection valve intermittently in a vehicle stop state upon receiving the forcible learning command signal from the issuance means.
The fuel injection amount learning control apparatus according to any one of claims 1 to 9, wherein the engine is a diesel engine.
The fuel injection amount learning control apparatus according to claim 10, wherein the learning value of the fuel injection amount is acquired by performing the learning operation for a plurality of levels of common rail pressures and for respective cylinders.