JP7609019B2 - Control device for internal combustion engine - Google Patents

Description

This invention relates to a control device for an internal combustion engine.

The internal combustion engine disclosed in Patent Document 1 has cylinders, spark plugs that ignite the cylinders, and a knock sensor that detects vibrations in the internal combustion engine. A control device that controls this internal combustion engine detects the occurrence of knocking based on the output of the knock sensor. In addition, if knocking occurs in the internal combustion engine, this control device corrects the ignition timing by retarding it.

JP 2019-148185 A

Some internal combustion engines are equipped with both a turbocharger and an EGR passage. The turbocharger has a compressor wheel located in the intake passage and a turbine wheel located in the exhaust passage. The turbine wheel rotates according to the flow of the exhaust air. The compressor wheel rotates together with the turbine wheel to supercharge the intake air. The EGR passage connects the upstream portion of the exhaust passage as seen from the turbine wheel to the downstream portion of the intake passage as seen from the compressor wheel. In this configuration, the EGR passage recirculates exhaust gas from the upstream portion of the exhaust passage as seen from the turbine wheel back to the intake passage.

In an internal combustion engine such as the one described above, when the accelerator pedal is depressed, the engine speed of the internal combustion engine increases relatively quickly, but there is a time lag before the rotation speed of the turbine wheel increases. In a transient state where the rotation speed of the turbine wheel has not yet fully increased, the amount of supercharging by the compressor wheel is small, and the amount of exhaust gas is also small accordingly. In such a situation, the amount of exhaust gas recirculated to the intake passage through the EGR passage, i.e., the amount of inert gas introduced into the cylinder, may not reach the intended amount. In this case, even if the ignition timing is retarded as in the technology of Patent Document 1, the amount of retardation may be insufficient, and knocking may occur.

The control device for an internal combustion engine for solving the above problems is applied to an internal combustion engine having a cylinder in which a mixture of intake air and fuel is burned, an ignition plug for igniting the mixture in the cylinder, an intake passage and an exhaust passage connected to the cylinder, a turbine wheel located in the exhaust passage, and a supercharger having a compressor wheel located in the intake passage and rotating integrally with the turbine wheel, and an EGR passage connecting a portion of the exhaust passage upstream from the turbine wheel and a portion of the intake passage downstream from the compressor wheel, and the control device includes a feedback correction value calculated according to the presence or absence of knocking in the internal combustion engine, and a feedback correction value calculated according to the presence or absence of knocking in the internal combustion engine. An ignition timing correction process can be executed to correct the ignition timing of the spark plug based on a knocking learning value calculated as a value that suppresses an increase in the absolute value of the feedback correction value. In the ignition timing correction process, when knocking occurs in the internal combustion engine, the feedback correction value is updated so that the ignition timing is retarded by a certain angle, and when the feedback correction value becomes a value that is more retarded than a predetermined specified value, the knocking learning value is updated to the side that retards the ignition timing. During an increase in engine speed, the amount of change to the retard side per update of the feedback correction value is increased compared to when the engine speed is in a steady state, and updating of the knocking learning value is prohibited.

In the above configuration, while the engine speed is increasing, the amount of change to the retard side per update of the feedback correction value is increased. In this case, the ignition timing of the spark plug can be quickly retarded. Therefore, even if the amount of exhaust gas recirculating to the intake passage via the EGR passage is small, the occurrence of knocking can be suppressed. In addition, in the above configuration, the update of the knocking learning value is prohibited in conjunction with the increase in the amount of change per update of the feedback correction value. Therefore, the knocking learning value is not updated to a value corresponding to a situation in which knocking is likely to occur.

FIG. 1 is a schematic diagram of an internal combustion engine. FIG. 2 is a diagram for explaining a situation in which condition (b) is met. FIG. 3 is a flowchart showing the procedure of the information designation process.

Hereinafter, an embodiment of a control device for an internal combustion engine will be described with reference to the drawings.
1, the vehicle 300 has an internal combustion engine 100. The internal combustion engine 100 is a drive source of the vehicle 300.

The internal combustion engine 100 has an engine body 100A and a knock sensor 19. The engine body 100A has a plurality of cylinders 11. Note that FIG. 1 shows only one of the plurality of cylinders 11. The cylinder 11 is a space partitioned by the engine body 100A. In the cylinder 11, a mixture of intake air and fuel is burned. The knock sensor 19 is attached to the engine body 100A. The knock sensor 19 detects the amount of vibration KN of the engine body 100A.

The internal combustion engine 100 has a plurality of pistons 12, a plurality of connecting rods 18, a crankshaft 15, and a crank angle sensor 13. As with the cylinders 11, FIG. 1 shows only one of the plurality of pistons 12. The same is true for the connecting rod 18. A piston 12 is provided for each cylinder 11. A connecting rod 18 is also provided for each cylinder 11. The piston 12 is located in the cylinder 11. The piston 12 can reciprocate within the cylinder 11. The connecting rod 18 connects the piston 12 to the crankshaft 15. The crankshaft 15 rotates in response to the reciprocating motion of the piston 12. The crank angle sensor 13 is located near the crankshaft 15. The crank angle sensor 13 detects the rotational position Scr of the crankshaft 15.

The internal combustion engine 100 has multiple spark plugs 14. Note that FIG. 1 shows only one of the multiple spark plugs 14. A spark plug 14 is provided for each cylinder 11. The tip of the spark plug 14 is located inside the cylinder 11. The spark plug 14 ignites the air-fuel mixture in the cylinder 11.

The internal combustion engine 100 has an intake passage 20 and an exhaust passage 30. The intake passage 20 is a passage for introducing intake air into each cylinder 11. The intake passage 20 is connected to each cylinder 11. The exhaust passage 30 is a passage for discharging exhaust gas from each cylinder 11. The exhaust passage 30 is connected to each cylinder 11.

The internal combustion engine 100 has a turbocharger 40. The turbocharger 40 has a compressor wheel 41 and a turbine wheel 42. The turbine wheel 42 is located in the exhaust passage 30. The turbine wheel 42 rotates with the flow of the exhaust gas. The compressor wheel 41 is located in the intake passage 20. The compressor wheel 41 rotates together with the turbine wheel 42. In response to the rotation of the compressor wheel 41, the compressor wheel 41 supercharges the intake air.

The internal combustion engine 100 has an airflow meter 22, an intake pressure sensor 27, an intercooler 23, a throttle valve 24, and a plurality of fuel injection valves 26. The airflow meter 22 is located upstream of the compressor wheel 41 in the intake passage 20. The airflow meter 22 detects the flow rate GA of the intake air flowing through the intake passage 20. The intake pressure sensor 27 is located downstream of the compressor wheel 41 in the intake passage 20. The intake pressure sensor 27 detects the intake pressure (hereinafter referred to as the supercharging pressure) PR of the portion of the intake passage 20 where the intake pressure sensor 27 is located. The intercooler 23 is located downstream of the intake pressure sensor 27 in the intake passage 20. The intercooler 23 cools the intake air. The throttle valve 24 is located downstream of the intercooler 23 in the intake passage 20. The throttle valve 24 opens and closes the intake passage 20. A fuel injection valve 26 is provided for each cylinder 11. The fuel injection valve 26 injects fuel into the cylinder 11 through the intake passage 20. Note that FIG. 1 shows only one of the multiple fuel injection valves 26.

The internal combustion engine 100 has a bypass passage 44 and a wastegate valve (hereinafter referred to as WGV) 46. The bypass passage 44 connects the upstream side and downstream side of the exhaust passage 30 when viewed from the turbine wheel 42. In other words, the bypass passage 44 is a passage that bypasses the turbine wheel 42. The WGV 46 is located in the bypass passage 44. The WGV 46 opens and closes the bypass passage 44.

The internal combustion engine 100 has an EGR passage 50 and an EGR valve 53. The EGR passage 50 connects an upstream portion of the exhaust passage 30 as viewed from the turbine wheel 42 to a downstream portion of the intake passage 20 as viewed from the throttle valve 24. The EGR passage 50 is a passage for recirculating exhaust gas to the intake passage 20. The EGR valve 53 is located midway through the EGR passage 50. The EGR valve 53 opens and closes the EGR passage 50.

The vehicle 300 has an accelerator sensor 301 and a vehicle speed sensor 302. The accelerator sensor 301 detects the amount of operation of the accelerator pedal of the vehicle 300 as an accelerator operation amount ACCP. The vehicle speed sensor 302 detects the traveling speed of the vehicle 300 as a vehicle speed SP.

<General configuration of the control device>
The vehicle 300 has a control device 110. The control device 110 may be configured as one or more processors that execute various processes according to a computer program (software). The control device 110 may be configured as one or more dedicated hardware circuits, such as an application specific integrated circuit (ASIC), that execute at least a part of the various processes, or a circuit (circuitry) including a combination thereof. The processor includes a CPU and memories such as RAM and ROM. The memory stores program code or instructions configured to cause the CPU to execute the processes. The memory, i.e., computer-readable medium, includes any available medium that can be accessed by a general-purpose or dedicated computer. The control device 110 has a storage device that is an electrically rewritable non-volatile memory.

The control device 110 receives detection signals from various sensors mounted on the vehicle 300. Specifically, the control device 110 receives detection signals for the following parameters:
The flow rate GA of the intake air detected by the air flow meter 22
The boost pressure PR detected by the intake pressure sensor 27
The rotational position Scr of the crankshaft 15 detected by the crank angle sensor 13
The vibration amount KN of the engine body 100A detected by the knock sensor 19
Accelerator operation amount ACCP detected by the accelerator sensor 301
Vehicle speed SP detected by the vehicle speed sensor 302
The control device 110 calculates parameters indicating the engine operating state at any time based on detection signals received from various sensors. Specifically, the control device 110 calculates the engine speed NE based on the rotational position Scr of the crankshaft 15. The control device 110 also calculates the engine load factor KL based on the engine speed NE and the intake air flow rate GA.

The control device 110 also constantly determines whether knocking has occurred in the internal combustion engine 100 based on the vibration amount KN of the engine body 100A detected by the knock sensor 19. If the vibration amount KN of the engine body 100A is within the allowable range, the control device 110 determines that knocking has not occurred. On the other hand, if the vibration amount KN of the engine body 100A exceeds the allowable range, the control device 110 determines that knocking has occurred.

The control device 110 controls various parts of the internal combustion engine 100. For example, the control device 110 controls the WGV 46. When the required output value for the internal combustion engine 100 is high, such as when the vehicle 300 is accelerating, the control device 110 closes the WGV 46. The control device 110 calculates the required output value for the internal combustion engine 100 based on the accelerator operation amount ACCP. The control device 110 also controls the EGR valve 53. The control device 110 basically opens the EGR valve 53 except when the engine speed NE or the engine load factor KL is considerably small. For example, the higher the engine speed NE or the higher the engine load factor KL, the larger the opening of the EGR valve 53 becomes.

<Ignition timing correction process>
The control device 110 has an ignition control unit 112 as one of a plurality of functional units that control the internal combustion engine 100. The ignition control unit 112 controls the ignition plug 14. As a process for this purpose, the ignition control unit 112 is capable of executing an ignition timing correction process.

The ignition timing correction process has a main process and a sub-process. The main process is a process for calculating the final ignition timing AFIN, which is the target value for the ignition timing of the spark plug 14. The final ignition timing AFIN is an ignition timing that is set as far forward as possible while suppressing the occurrence of knocking.

A method for calculating the final ignition timing AFIN will now be described.
The ignition control unit 112 calculates the final ignition timing AFIN based on the following equations (1) to (3). In this embodiment, the ignition timing is expressed as an advance amount of the rotational position Scr of the crankshaft 15 with respect to the compression top dead center of the cylinder 11 to be ignited.

As shown in equation (1), the final ignition timing AFIN is a value obtained by subtracting the retard amount AKNK from the base ignition timing ABASE.
AFIN=ABASE-AKNK…(1)
In formula (1), the base ignition timing ABASE is the more retarded value of the MBT ignition timing AMBT and the knock limit ignition timing AKNOK. The MBT ignition timing AMBT is the ignition timing that can obtain the maximum torque in the current engine operating state. The knock limit ignition timing AKNOK is the advance limit timing of the ignition timing that can keep the vibration amount KN of the engine body 100A within the allowable range under the best assumed conditions when using a high octane fuel with a high knock limit.

The retard amount AKNK in formula (1) is the maximum retard amount AKMAX minus the knocking learning value AGKNK plus the feedback correction value AKCS, as shown in formula (2). The larger the retard amount AKNK, the more retarded the final ignition timing AFIN becomes, making knocking less likely to occur.

AKNK=AKMAX-AGKNK+AKCS...(2)
In the formula (2), the feedback correction value AKCS is a term for quickly correcting the final ignition timing AFIN depending on whether or not knocking occurs. When the feedback correction value AKCS is negative, the retard amount AKNK As a result, the final ignition timing AFIN calculated from the above formula (1) is corrected to a timing further advanced. On the other hand, when the value of the feedback correction value AKCS becomes positive, In this case, the retard amount AKNK is increased. As a result, the final ignition timing AFIN calculated from the above formula (1) is corrected to a more retarded timing. The knocking learning value AGKNK is , a term for suppressing an increase in the absolute value of the feedback correction value AKCS.

The maximum retard amount AKMAX in equation (2) is a value obtained by subtracting the most retarded ignition timing AKMF from the base ignition timing ABASE, as shown in the following equation (3).
AKMAX=ABASE-AKMF…(3)
In equation (3), the most retarded ignition timing AKMF is the maximum advancement timing of the ignition timing that can keep the vibration amount KN of the engine body 100A within the allowable range even under the worst-case conditions assumed when using a low-octane fuel with a low knock limit.

The ignition control unit 112 calculates the final ignition timing AFIN using the above formulas (1) to (3) once per execution of the main process. The ignition control unit 112 repeats such main process at a predetermined control period while the internal combustion engine 100 is operating. After calculating the final ignition timing AFIN, the ignition control unit 112 controls the spark plug 14 so that ignition occurs at the final ignition timing AFIN.

When calculating the final ignition timing AFIN, the ignition control unit 112 calculates the base ignition timing ABASE used in equations (1) and (3) based on the latest engine speed NE and the latest engine load factor KL. Similarly, the ignition control unit 112 calculates the knock limit ignition timing AKNOK used in equation (1) and the most retarded ignition timing AKMF used in equation (3) based on the latest engine speed NE and the latest engine load factor KL. For the knocking learning value AGKNK used in equation (2), the ignition control unit 112 uses the value calculated in a sub-process described below.

The ignition control unit 112 also updates the feedback correction value AKCS used in equation (2) each time the main process is executed. That is, the ignition control unit 112 holds the latest value of the feedback correction value AKCS that was previously calculated. The ignition control unit 112 updates the value of the feedback correction value AKCS by performing calculation processing on the feedback correction value AKCS that is currently held. The ignition control unit 112 then uses the updated value to calculate the final ignition timing AFIN.

The ignition control unit 112 updates the feedback correction value AKCS depending on whether or not knocking has occurred. When knocking has not occurred in the internal combustion engine 100, the ignition control unit 112 subtracts the positive advance update value Z1 from the currently held feedback correction value AKCS to set the latest feedback correction value AKCS. In other words, the control device 110 updates the feedback correction value AKCS so that the final ignition timing AFIN is advanced by a certain angle. The above advance update value Z1 is determined in advance, for example, through experiments or simulations, as a value suitable for gradually advancing the final ignition timing AFIN while suppressing the occurrence of knocking.

On the other hand, when knocking occurs in the internal combustion engine 100, the ignition control unit 112 adds the positive retardation update value Z2 to the currently held feedback correction value AKCS, and sets the result as the latest feedback correction value AKCS. In other words, the ignition control unit 112 updates the feedback correction value AKCS so that the final ignition timing AFIN is retarded by a certain angle. There are two retardation update values Z2: a normal update value Z2N and an acceleration update value Z2T. The ignition control unit 112 selects one of the normal update value Z2N and the acceleration update value Z2T as the retardation update value Z2 based on the use prohibition flag F1 described below.

<Details of retardation update value>
When the accelerator pedal is depressed to accelerate the vehicle 300, the engine speed NE increases relatively quickly. On the other hand, the turbine wheel 42, which rotates according to the flow of exhaust gas, requires time to start rotating. Therefore, in a situation where the engine speed NE starts to increase at the beginning of acceleration, the actual rotation speed TWR of the turbine wheel 42 is lower than the steady rotation speed TWK, which is a rotation speed according to the current engine operating state. Here, it is assumed that the current engine speed NE is a first value, and the current engine load factor KL is a second value. The steady rotation speed TWK is the rotation speed of the turbine wheel 42 when it is assumed that the engine speed NE is constant at a first value in the time change of the engine speed NE, and the engine load factor KL is constant at a second value in the time change of the engine load factor KL. Now, in a situation where the actual rotation speed TWR of the turbine wheel 42 is lower than the steady rotation speed TWK, the amount of supercharging by the compressor wheel 41 is small, and the amount of exhaust gas is also small accordingly. As a result, the amount of exhaust gas recirculated to the intake passage 20 through the EGR passage 50, i.e., the amount of inert gas introduced into the cylinder 11, may not reach the amount originally targeted, which may result in knocking. The acceleration update value Z2T is a value required to quickly retard the final ignition timing AFIN and suppress the occurrence of knocking when the actual rotation speed TWR of the turbine wheel 42 is lower than the steady rotation speed TWK by a determination value Q or more. The acceleration update value Z2T is determined in advance, for example, through experiments or simulations. The definition of the determination value Q will be described later.

The normal update value Z2N is a value suitable for suppressing the occurrence of knocking under a situation where the actual rotation speed TWR of the turbine wheel 42 is approximately equal to the steady rotation speed TWK. The normal update value Z2N is smaller than the acceleration update value Z2T. The normal update value Z2N is determined in advance, for example, by experiment or simulation. The normal update value Z2N is effective under a situation where the engine operating state is approximately steady, such as a period when the vehicle 300 transitions from acceleration to constant speed running, and a period after that when the vehicle 300 runs at a constant speed. Here, the value obtained by subtracting the actual rotation speed TWR of the turbine wheel 42 from the steady rotation speed TWK corresponding to the current engine operating state is called the rotation speed difference ΔTW. The above-mentioned judgment value Q is determined, for example, by experiment or simulation, as the minimum value of the rotation speed difference ΔTW at which the occurrence of knocking cannot be dealt with by using the normal update value Z2N in suppressing the occurrence of knocking.

Of the three driving states of the vehicle 300, namely, accelerating, cruising at a constant speed, and decelerating, ignition is not necessarily required during deceleration because, for example, fuel cut is performed. Of the remaining two driving states, accelerating and cruising at a constant speed, the above acceleration update value Z2T and normal update value Z2N are determined based on the correspondence between the actual rotation speed TWR and steady-state rotation speed TWK of the turbine wheel 42 for the initial acceleration period and the remaining periods. In the following description, the actual rotation speed TWR of the turbine wheel 42 may also be referred to as the current rotation speed TWR of the turbine wheel 42 depending on the context.

<Sub-processing>
The sub-processing of the ignition timing correction process is a process for updating the knocking learning value AGKNK. The ignition control unit 112 continues to execute the sub-processing during the operation of the internal combustion engine 100, on the condition that an execution prohibition flag F2 described later is off. In the sub-processing, the ignition control unit 112 updates the knocking learning value AGKNK when the absolute value of the feedback correction value AKCS becomes large to a certain extent. That is, the ignition control unit 112 holds the latest value of the previously calculated knocking learning value AGKNK, similar to the feedback correction value AKCS. Then, when the state in which the absolute value of the feedback correction value AKCS is larger than the specified value A continues for a determination period or more, the ignition control unit 112 performs an arithmetic process on the currently held knocking learning value AGKNK so as to reduce the absolute value of the feedback correction value AKCS.

More specifically, when the feedback correction value AKCS continues to be a positive value with an absolute value greater than the specified value A, that is, when knocking is likely to occur, the ignition control unit 112 updates the knocking learning value AGKNK as follows. That is, as shown in equation (4), the ignition control unit 112 subtracts a positive change value B from the currently held value of the knocking learning value AGKNKp. At the same time, the ignition control unit 112 also subtracts the change value B from the currently held value of the feedback correction value AKCSp. As a result, the absolute value of the feedback correction value AKCS after subtraction becomes a value equal to or less than the specified value A.

AKNK=AKMAX-(AGKNKp-B)+(AKCSp-B)...(4)
As described above, by subtracting the same value B from both the knocking learning value AGKNKp and the feedback correction value AKCSp, the value of the retardation amount AKNK is changed from the value before the subtraction as shown in the formula (4). That is, even if the knocking learning value AGKNK is updated, the value of the final ignition timing AFIN does not change.

Here, as shown in the above formulas (1) and (2), when the feedback correction value AKCS becomes larger, the final ignition timing AFIN becomes more retarded through the retard amount AKNK. In other words, the larger the feedback correction value AKCS, the more retarded the ignition timing is. Therefore, updating the knocking learning value AGKNK when the feedback correction value AKCS is larger than the positive specified value A as described above means updating the knocking learning value AGKNK when the feedback correction value AKCS is more retarded than the specified value A. When updating the knocking learning value AGKNK, the change value B is subtracted from the currently held knocking learning value AGKNKp as shown in formula (4). This contributes to increasing the retard amount AKNK in consideration of the sign in formula (4). In other words, the ignition control unit 112 updates the knocking learning value AGKNK to the side where the final ignition timing AFIN is retarded. In this way, in the sub-processing, when the feedback correction value AKCS is more retarded than the specified value A, the ignition control unit 112 updates the knocking learning value AGKNK to the side where the final ignition timing AFIN is retarded.

On the other hand, when the feedback correction value AKCS remains negative and its absolute value is greater than the specified value A, that is, when knocking is unlikely to occur, the ignition control unit 112 updates the knocking learning value AGKNK as follows. That is, the ignition control unit 112 adds the above-mentioned change value B to the currently held values of the knocking learning value AGKNK and the feedback correction value AKCS. As a result, the absolute value of the feedback correction value AKCS after the addition becomes equal to or less than the specified value A. By adding the same value B to both the knocking learning value AGKNK and the feedback correction value AKCS in the above-mentioned manner, the retardation amount AKNK and the value of the final ignition timing AFIN do not change from the values before the addition.

The specified value A is predefined as the maximum allowable absolute value of the feedback correction value AKCS. The change value B is predefined as a value that can make the absolute value of the feedback correction value AKCS sufficiently smaller than the specified value A. The judgment period is predefined as the length of time during which the state in which the feedback correction value AKCS is greater than the specified value A is considered to be continuous rather than momentary. The values of these parameters are set, for example, through experiments or simulations.

<Information specification processing>
The control device 110 has an information designation unit 114 as one of a plurality of functional units that control the internal combustion engine 100. The information designation unit 114 is capable of executing information designation processing for designating information required in the ignition timing correction processing. More specifically, the information designation processing is processing for designating the retardation update value Z2 to be used in the main processing, and for designating whether or not the sub-processing is to be executed.

In the information designation process, the information designation unit 114 switches the information to be designated depending on whether or not both of the following two conditions (a) and (b) are satisfied.
(a) In terms of the change over time of the engine speed NE, the engine speed NE is increasing.

(b) The rotational speed difference ΔTW relating to the turbine wheel 42 is equal to or greater than the judgment value Q.
As described above, in the early stage of acceleration, the actual rotation speed TWR of the turbine wheel 42 becomes lower than the steady rotation speed TWK, and the rotation speed difference ΔTW becomes large. Therefore, the situation in which both of the two conditions (a) and (b) are satisfied substantially corresponds to the situation in the early stage of acceleration of the vehicle 300. Here, as shown in FIG. 2, consider an orthogonal coordinate system in which the actual rotation speed TWR of the turbine wheel 42 is on the horizontal axis and the steady rotation speed TWK is on the vertical axis. In this orthogonal coordinate system, a straight line with a slope of "1" and passing through the origin is called a first straight line D1. This first straight line D1 indicates that the actual rotation speed TWR of the turbine wheel 42 and the steady rotation speed TWK are the same value. For example, as shown by a black circle V1 in FIG. 2, when a combination of the actual rotation speed TWR of the turbine wheel 42 and the steady rotation speed TWK according to the current engine operating state is located in an area near the first straight line D1, the condition (b) is not satisfied. The line with a slope of "1" and an intercept of "-Q" is referred to as the second line D2. For example, as shown by a black circle V2 in Fig. 2, when a combination of the actual rotation speed TWR of the turbine wheel 42 and the steady rotation speed TWK according to the current engine operating state is located in an area DA between the second line D2 and the horizontal axis, condition (ii) is satisfied.

The information designation unit 114 prestores a first map as information for calculating the steady rotation speed TWK corresponding to the current engine operating state. Specifically, the first map represents the correspondence between the engine rotation speed NE, the engine load factor KL, and the steady rotation speed TWK. That is, the first map represents the steady rotation speed TWK for each of various combinations of the engine rotation speed NE and the engine load factor KL. The first map is created, for example, through an experiment or a simulation.

The information designation unit 114 also prestores a second map as information for calculating the actual rotation speed TWR of the turbine wheel 42. Specifically, the second map represents the correspondence between the intake flow rate GA, the boost pressure PR, and the actual rotation speed TWR of the turbine wheel 42. That is, the second map represents the actual rotation speed TWR of the turbine wheel 42 for various combinations of the intake flow rate GA and the boost pressure PR. The second map is created, for example, through an experiment or a simulation.

Now, when both of the above two conditions (a) and (b) are met, the information designation unit 114 designates the use of the acceleration update value Z2T as the retardation update value Z2. In other words, the amount of change to the retardation side per update of the feedback correction value AKCS is made larger than when the engine operating state is in a steady state. In addition, when both of the above two conditions (a) and (b) are met, the information designation unit 114 prohibits the execution of the sub-processing. In other words, the information designation unit 114 prohibits the update of the knocking learning value AGKNK.

On the other hand, when at least one of the above two conditions (a) and (b) is not satisfied, the information specification unit 114 specifies the use of the normal update value Z2N as the retardation update value Z2. Also, when at least one of the above two conditions (a) and (b) is not satisfied, the information specification unit 114 allows the execution of the sub-processing.

<Specific processing procedure of information designation processing>
The information designation unit 114 repeats the information designation process at a predetermined control period while the internal combustion engine 100 is operating. As shown in FIG. 3, when the information designation unit 114 starts the information designation process, it first executes the process of step S10. In step S10, the information designation unit 114 judges whether the engine rotation speed NE is increasing. In making this judgment, the information designation unit 114 refers to the history of the engine rotation speed NE. Regarding the engine rotation speed NE at two consecutive timings, a state in which a newer value is higher than an older value is called an increasing state. In the history of the engine rotation speed NE, if the increasing state continues for a specified period or more, the information designation unit 114 judges that the engine rotation speed NE is increasing, and if not, judges that the engine rotation speed NE is not increasing. The specified period is predetermined as a length of time during which the increasing state can be considered to be continuous and not instantaneous. If the engine rotation speed NE is not increasing (step S10: NO), the information designation unit 114 advances the process to step S130.

In step S130, the information designation unit 114 designates the normal update value Z2N as the retardation update value Z2 used to update the feedback correction value AKCS in the main process. As a specific process of this step S130, the information designation unit 114 sets the use prohibition flag F1, which is a flag indicating that the use of the normal update value Z2N is prohibited, to OFF. Here, the ignition control unit 112 is set to prioritize the use of the normal update value Z2N between the normal update value Z2N and the acceleration update value Z2T. Therefore, when the use prohibition flag F1 is OFF, the ignition control unit 112 selects the normal update value Z2N as the retardation update value Z2 in the main process. After setting the use prohibition flag F1 to OFF, the information designation unit 114 proceeds to step S140.

In step S140, the information specification unit 114 permits the execution of the sub-processing, i.e., the update of the knocking learning value AGKNK. As a specific process of this step S140, the information specification unit 114 sets the execution prohibition flag F2, which is a flag indicating that the execution of the sub-processing is prohibited, to OFF. After this, the information specification unit 114 temporarily ends the series of processes of the information specification process.

On the other hand, in step S10, if the engine rotation speed NE is increasing (step S10: YES), the information designation unit 114 advances the process to step S20.
In step S20, the information designation unit 114 determines whether the rotational speed difference ΔTW of the turbine wheel 42 is equal to or greater than the determination value Q. As a specific process of step S20, the information designation unit 114 first calculates a steady rotational speed TWK corresponding to the current engine operating state. That is, the information designation unit 114 refers to the latest engine rotational speed NE, the latest engine load factor KL, and the first map. Then, the information designation unit 114 calculates the steady rotational speed TWK corresponding to the latest engine rotational speed NE and the latest engine load factor KL based on the first map. After that, the information designation unit 114 calculates the current rotational speed TWR of the turbine wheel 42. That is, the information designation unit 114 refers to the latest intake flow rate GA, the latest supercharging pressure PR, and the second map. Then, the information designation unit 114 calculates the rotational speed TWR of the turbine wheel 42 corresponding to the latest intake flow rate GA and the latest supercharging pressure PR based on the second map. Thereafter, the information designation unit 114 calculates the rotation speed difference ΔTW by taking the difference between the two parameters calculated as described above. Then, the information designation unit 114 compares the rotation speed difference ΔTW with the judgment value Q. If the rotation speed difference ΔTW is less than the judgment value Q (step S20: NO), the information designation unit 114 advances the process to step S130.

On the other hand, if the rotation speed difference ΔTW is equal to or greater than the determination value Q (step S20: YES), the information designation unit 114 advances the process to step S30.
In step S30, the information designation unit 114 designates the acceleration update value Z2T as the retardation update value Z2 used to update the feedback correction value AKCS in the main processing. As a specific process of this step S30, the information designation unit 114 sets the above-mentioned use prohibition flag F1, which is a flag indicating prohibition of use of the normal update value Z2N, to ON. In this case, the ignition control unit 112 selects the acceleration update value Z2T as the retardation update value Z2 in the main processing. After setting the use prohibition flag F1 to ON, the information designation unit 114 proceeds to the process of step S40.

In step S40, the information designation unit 114 prohibits the execution of the sub-processing, i.e., prohibits the update of the knocking learning value AGKNK. As a specific process of this step S40, the information designation unit 114 sets the execution prohibition flag F2, which is a flag indicating that the execution of the sub-processing is prohibited, to ON. After this, the information designation unit 114 temporarily ends the series of processes of the information designation process.

<Operation of the embodiment>
Assume that the accelerator pedal is depressed. Then, the engine rotation speed NE starts to increase in order to accelerate the vehicle 300 (step S10: YES). At this time, the rotation speed difference ΔTW becomes large due to the fact that the turbine wheel 42 requires time to start rotating (step S20: YES). Then, the information designation unit 114 turns on the use prohibition flag F1 of the normal update value Z2N in order to instruct the use of the acceleration update value Z2T as the retardation update value Z2 (step S30). In response to this, the ignition control unit 112 updates the feedback correction value AKCS using the acceleration update value Z2T. In this case, the amount of change to the retardation side per update of the feedback correction value AKCS becomes larger than when the feedback correction value AKCS is updated using the normal update value Z2N under the condition that the engine rotation speed NE is in a steady state. In addition, in response to the instruction to use the acceleration update value Z2T as the retardation update value Z2, the information designation unit 114 turns on the execution prohibition flag F2 to prohibit updating of the knocking learning value AGKNK (step S40). In response to this, the ignition control unit 112 stops execution of the sub-processing of the ignition timing correction process, i.e., stops updating of the knocking learning value AGKNK.

Effects of the embodiment
When the rotational speed difference ΔTW of the turbine wheel 42 is large at the beginning of acceleration of the vehicle 300, the amount of supercharging by the compressor wheel 41 is small, and the amount of exhaust gas is also small accordingly. In such a situation, the amount of exhaust gas recirculated to the intake passage 20 through the EGR passage 50, i.e., the amount of inert gas introduced into the cylinder 11, may not reach the originally targeted amount.

Therefore, in this embodiment, when the vehicle 300 is in the early stage of acceleration and the rotational speed difference ΔTW is large, the amount of change to the retard side per update of the feedback correction value AKCS is increased. In this case, the final ignition timing AFIN can be quickly corrected to retard. Therefore, even if the amount of exhaust gas recirculating to the intake passage 20 via the EGR passage 50 is small, the occurrence of knocking can be suppressed. Furthermore, in this embodiment, in conjunction with increasing the amount of change per update of the feedback correction value AKCS, updating of the knocking learning value AGKNK is prohibited. Therefore, the knocking learning value AGKNK is not updated to a value corresponding to a situation in which knocking is likely to occur.

<Example of change>
This embodiment can be modified as follows: This embodiment and the following modifications can be combined with each other to the extent that there is no technical contradiction.

- The content of the sub-processing, i.e., the method of updating the knocking learning value AGKNK, is not limited to the example of the above embodiment. For example, the conditions for updating the knocking learning value AGKNK may be changed from the example of the above embodiment. For example, a moving average is applied to the time series of the feedback correction value AKCS. Then, when the absolute value of the time series value after the moving average is applied becomes larger than the specified value A, the knocking learning value AGKNK may be updated. If the knocking learning value AGKNK can be updated to the side where the final ignition timing AFIN is retarded when the feedback correction value AKCS becomes a value on the retard side of the predetermined specified value A, the method of updating the knocking learning value AGKNK does not matter. The specified value A may be set to an appropriate value according to the method of updating the knocking learning value AGKNK. When the change value B is used as in the above embodiment, the change value B may also be set to an appropriate value according to the method of updating the knocking learning value AGKNK.

- The method of updating the feedback correction value AKCS is not limited to the example of the above embodiment. For example, the change amount to the retard side per update of the feedback correction value AKCS, i.e., the retard update value Z2, may be changed each time depending on the magnitude of the vibration amount KN of the engine body 100A. Even in this case, it is sufficient that a different retard update value Z2 is pre-determined depending on the magnitude of the vibration amount KN. In other words, the retard update value Z2 does not always have to be a constant fixed value, and if the feedback correction value AKCS is updated with the same change amount under the same circumstances, it can be said that the feedback correction value AKCS is updated so as to retard the final ignition timing AFIN by a certain angle. The method of updating the feedback correction value AKCS may be any method that can update the feedback correction value AKCS so that the final ignition timing AFIN is retarded by a certain angle when knocking occurs in the internal combustion engine 100.

- As in the above modified example, when the retardation update value Z2 is set to be variable, both the acceleration update value Z2T and the normal update value Z2N may be set to be variable, or only one of them may be set to be variable. For example, when both the acceleration update value Z2T and the normal update value Z2N are set to be variable, it is sufficient that the acceleration update value Z2T and the normal update value Z2N are determined so that the minimum value that the acceleration update value Z2T can take is greater than the maximum value that the normal update value Z2N can take. From a similar perspective, when only one of the acceleration update value Z2T and the normal update value Z2N is set to be variable, it is sufficient that these values are determined so that the acceleration update value Z2T is greater than the normal update value Z2N.

- In the information specification process, the process of step S20 relating to condition (ii) may be omitted. Regardless of the magnitude of the rotational speed difference ΔTW, while the engine rotational speed NE is increasing, the amount of exhaust gas recirculated to the intake passage 20 via the EGR passage 50 may decrease. Therefore, in order to suppress the occurrence of knocking, it is also effective to use the acceleration update value Z2T to update the feedback correction value AKCS and to prohibit the knocking learning value AGKNK only under condition (i) that the engine rotational speed NE is increasing.

The method of determining that the engine speed NE is increasing is not limited to the example of the above embodiment. As long as it can be determined that the engine speed NE is continuously increasing, any method of determination is acceptable. The engine speed NE increasing corresponds to the vehicle 300 accelerating. Therefore, it may be determined that the engine speed NE is increasing when the acceleration of the vehicle 300 is greater than zero. In this case, an acceleration sensor may be attached to the vehicle 300. Furthermore, when determining that the engine speed NE is increasing, the determination may be made based on the accelerator operation amount ACCP and the vehicle speed SP instead of or in addition to the engine speed NE.

- In the above embodiment, the steady-state rotation speed TWK was defined as the rotation speed of the turbine wheel 42 when both the engine rotation speed NE and the engine load factor KL are steady. Alternatively, the steady-state rotation speed TWK may be defined by considering only the engine rotation speed NE, without considering the engine load factor KL. In other words, the steady-state rotation speed TWK may be defined as the rotation speed of the turbine wheel 42 when only the engine rotation speed NE is steady.

- The method of calculating the final ignition timing AFIN is not limited to the example of the above embodiment. The final ignition timing AFIN may be corrected based on the feedback correction value AKCS and the knocking learning value AGKNK.

The configuration of the internal combustion engine 100 is not limited to the example of the above embodiment. The internal combustion engine 100 only needs to have a cylinder 11, an ignition plug 14, an intake passage 20, an exhaust passage 30, a turbocharger 40, and an EGR passage 50. For example, a rotation sensor that detects the rotational position of the turbine wheel 42 may be attached near the turbine wheel 42. The actual rotational speed TWR of the turbine wheel 42 may then be calculated based on the detection result of the rotation sensor.

The configuration of the vehicle 300 is not limited to the example of the above embodiment. For example, the vehicle 300 may have a motor generator in addition to the internal combustion engine 100 as a drive source for the vehicle 300.

REFERENCE SIGNS LIST 11 cylinder 14 spark plug 20 intake passage 30 exhaust passage 40 supercharger 41 compressor wheel 42 turbine wheel 50 EGR passage 100 internal combustion engine 110 control device 112 ignition control unit 114 information designation unit

Claims (1)

A cylinder in which an intake air/fuel mixture is burned;
A spark plug for igniting an ignition in the cylinder;
an intake passage and an exhaust passage connected to the cylinder;
a turbocharger including a turbine wheel located in the exhaust passage and a compressor wheel located in the intake passage and rotating integrally with the turbine wheel;
an EGR passage connecting a portion of the exhaust passage upstream of the turbine wheel and a portion of the intake passage downstream of the compressor wheel;
The present invention is applied to an internal combustion engine having
an ignition timing correction process can be executed to correct the ignition timing of the spark plug based on a feedback correction value calculated in accordance with the occurrence or absence of knocking in the internal combustion engine and a knocking learning value calculated as a value for suppressing an increase in the absolute value of the feedback correction value,
In the ignition timing correction process,
updating the feedback correction value so that the ignition timing is retarded by a certain angle when knocking occurs in the internal combustion engine;
When the feedback correction value becomes a value on the retard side relative to a predetermined specified value, the knocking learning value is updated to a side on which the ignition timing is retarded,
A control device for an internal combustion engine, wherein while an engine rotation speed is increasing, an amount of change to the retard side per update of the feedback correction value is increased compared to when the engine rotation speed is in a steady state, and updating of the knocking learning value is prohibited.

Images