JP2016125356A - Control device for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control device for a vehicle-mounted internal combustion engine mounted with a cylinder pressure sensor, for, in the stage that the output characteristics of the cylinder pressure sensor is changed with a change in thermal strain characteristics, avoiding a sensor error from being reflected to actuator control of the internal combustion engine as it is.SOLUTION: The control device for the internal combustion engine includes a cylinder pressure sensor 20 having a pressure sensitive part located in a cylinder of the internal combustion engine, and a control unit 40 for controlling actuators 16, 18 of the internal combustion engine 10 on the basis of the output of the cylinder pressure sensor 20. The control unit 40 determines whether a pre-stabilization change in the thermal strain characteristics of the pressure sensitive part 22 settles or not on the basis of the output of the cylinder pressure sensor 20, and keeps the reflection degree of the output of the cylinder pressure sensor 20 to the control, lower than the reflection degree in use after the determination is affirmed, until the determination is affirmed.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a control device for an internal combustion engine, and more particularly to a control device for an internal combustion engine that is suitable as a control device for a vehicle-mounted internal combustion engine equipped with an in-cylinder pressure sensor.

  Patent Document 1 discloses a control device for an internal combustion engine equipped with an in-cylinder pressure sensor. The in-cylinder pressure sensor has a pressure sensitive part exposed in the cylinder. The pressure sensitive part is provided with a diaphragm, and a shaft member is accommodated inside thereof. The pressure generated in the cylinder is transmitted into the cylinder pressure sensor by the diaphragm and the shaft member. The in-cylinder pressure sensor generates an output corresponding to the in-cylinder pressure Pc based on the magnitude of the force transmitted in this way.

  The in-cylinder pressure Pc is used for various controls related to the operation of the internal combustion engine, such as feedback control of the air-fuel ratio and ignition timing, torque fluctuation and misfire determination. For example, Patent Document 1 discloses an example in which air-fuel ratio feedback control is performed based on the in-cylinder pressure Pc.

JP 2014-020205 A Japanese Patent Laid-Open No. 08-068373 JP 2014-109212 A JP 2009-114916 A

  By the way, since the pressure sensitive part of the in-cylinder pressure sensor is exposed in the cylinder, it receives heat accompanying combustion. When the heat is received, the pressure-sensitive part is distorted, and the stress resulting from the distortion is transmitted to the inside of the in-cylinder pressure sensor. For this reason, in general, an error due to the influence of thermal distortion of the pressure sensitive portion is superimposed on the output of the in-cylinder pressure sensor.

  On the other hand, deposits generated with combustion adhere to the pressure-sensitive portion exposed in the cylinder. As the in-cylinder combustion is repeated, deposits are accumulated in the pressure sensitive part. This deposit has the effect of preventing heat reception of the pressure sensitive part. For this reason, as the deposit builds up, the pressure-sensitive portion becomes less susceptible to heat, and the error superimposed on the sensor output due to thermal strain is reduced.

  The above-described change relating to thermal strain appears particularly abruptly in a stage from the state where no deposit is attached to the pressure-sensitive portion until the deposit is slightly deposited. For this reason, the error that appears in the output of the in-cylinder pressure sensor due to thermal strain decreases rapidly after deposit deposition starts from the state where no deposit is attached to the pressure-sensitive portion, and eventually as deposit deposition proceeds. Converge to a stable value. Therefore, the output characteristic of the in-cylinder pressure sensor changes abruptly until the deposit amount of the pressure-sensitive portion reaches a certain level, and then converges to a stable state.

  In the above-described conventional control device, it is not considered that the above-described change appears in the output characteristics of the in-cylinder pressure sensor, and the in-cylinder pressure sensor is used regardless of how deposits are accumulated in the pressure-sensitive portion. Is always reflected in the air-fuel ratio feedback control under the same conditions. For this reason, in the said conventional apparatus, the deposit accumulation amount was inadequate and there existed a possibility that an inaccurate feedback control may be performed in the stage where the big change appears in the output characteristic of a cylinder pressure sensor.

  The present invention has been made to solve the above-described problems, and an error superimposed on the sensor output at the stage where the output characteristic of the in-cylinder pressure sensor changes due to a change in the thermal strain characteristic of the pressure-sensitive portion. However, it is an object of the present invention to provide a control device for an internal combustion engine that can be directly reflected in actuator control of the internal combustion engine.

In order to achieve the above object, a first invention is a control device for an internal combustion engine,
An in-cylinder pressure sensor having a pressure-sensitive portion located in a cylinder of the internal combustion engine;
A control unit for controlling the actuator of the internal combustion engine based on the output of the in-cylinder pressure sensor,
The control unit is
Based on the output of the in-cylinder pressure sensor, it is determined whether or not the pre-stabilization change of the thermal strain characteristics of the pressure sensitive part has been settled,
Until the determination is affirmed, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is suppressed to be lower than the degree of reflection used after the determination is affirmed.

The second invention is the first invention, wherein
The control unit is
Based on the output of the in-cylinder pressure sensor, the mass combustion ratio MFBax at a specific crank angle ax belonging to the region from the maximum in-cylinder pressure generation crank angle to the exhaust valve opening crank angle is calculated for the cylinder,
Determine whether the mass combustion rate MFBax has reached a stable value,
When it is determined that the mass combustion ratio MFBax has reached a stable value, it is determined that the pre-stabilization change of the thermal strain characteristics has been settled.

The third invention is the first invention, wherein
The control unit is
For the cylinder, it is determined whether or not the output Pcax of the in-cylinder pressure sensor at a specific crank angle ax belonging to the region from the maximum in-cylinder pressure generation crank angle to the exhaust valve opening crank angle has reached a stable value,
When it is determined that the output Pcax has reached a stable value, it is determined that the pre-stability change of the thermal strain characteristics has been settled.

  In a fourth aspect based on the second or third aspect, the specific crank angle ax belongs to a rear quarter region in a region from the maximum in-cylinder pressure generating crank angle to the exhaust valve opening crank angle. It is characterized by that.

In addition, a fifth invention is any one of the second to fourth inventions,
The control unit is
When the difference between the reference mass combustion rate MFBax0 to be generated at the specific crank angle ax and the mass combustion rate MFBax is equal to or less than a determination value, it is determined that the mass combustion rate MFBax has reached a stable value. And

In addition, a sixth invention is any one of the second to fourth inventions,
The control unit is
When the difference between the past mass combustion ratio MFBax1 obtained in the cylinder and the mass combustion ratio MFBax is equal to or less than a determination value, it is determined that the mass combustion ratio MFBax has reached a stable value,
The past mass combustion rate MFBax1 was calculated under conditions where it can be considered that the same mass combustion rate occurs at the specific crank angle ax as compared to the condition of the cylinder when the mass combustion rate MFBax was calculated. It is characterized by the latest value in the past.

In addition, a seventh invention is any one of the first inventions,
The control unit is
Calculate the amount of drift that appears in the output of the in-cylinder pressure sensor with the on / off switching of the fuel cut,
When the difference between the reference drift amount and the drift amount to be generated when no deposit is attached to the pressure-sensitive part under the condition in which the drift amount has occurred and the thermal strain characteristic is stable It is characterized by determining that the previous change has subsided.

Further, an eighth invention is any one of the first to seventh inventions,
The control of the actuator is
State determination based on the output of the in-cylinder pressure sensor;
A process for switching a command value calculation method for the actuator according to a result of the state determination;
Supplying the command value to the actuator, and
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has been affirmed, by prohibiting the execution of the state determination, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is determined. It is characterized by being kept lower than the degree of reflection used after affirmation.

Further, a ninth invention is any one of the first to eighth inventions,
The control of the actuator includes feedback control that reflects a correction amount based on the output of the in-cylinder pressure sensor in a command value for the actuator,
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has been affirmed, the feedback gain of the correction amount is made smaller than the gain used after the determination is affirmed, so that the in-cylinder pressure sensor The degree to which the output is reflected in the control is suppressed to be lower than the degree of reflection used after affirmation of the determination.

According to a tenth invention, in any one of the first to ninth inventions,
The control of the actuator includes feedback control that reflects a correction amount based on the output of the in-cylinder pressure sensor in a command value for the actuator,
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has settled is affirmed, by prohibiting the execution of the feedback control, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is determined. It is characterized by being kept lower than the degree of reflection used after affirmation.

The eleventh invention is the ninth or tenth invention,
The internal combustion engine is an internal combustion engine having a plurality of cylinders;
The in-cylinder pressure sensor is arranged for each cylinder,
The actuator includes a fuel injection valve arranged for each cylinder,
The feedback control includes air-fuel ratio feedback control that causes the fuel injection valve to inject a fuel injection amount to which an output of the in-cylinder pressure sensor is fed back,
The control unit is
For each cylinder, based on the output of the in-cylinder pressure sensor before it is determined that the pre-stability change of the thermal strain characteristic has been settled, the air-fuel ratio feature of the cylinder is grasped,
When it is determined that the pre-stability change of the thermal strain characteristic has been settled in at least one cylinder,
Based on the output of the in-cylinder pressure sensor of the stable cylinder for which the determination has been made, the injection correction amount of the stable cylinder is calculated,
From the relative relationship between the air-fuel ratio feature of an unstable cylinder that has not been determined that the pre-stability change of the thermal strain characteristic has subsided and the air-fuel ratio feature of the stable cylinder, the relative relationship between the injection correction amounts of both cylinders is specified. ,
By calculating the injection correction amount of the unstable cylinder by applying the relative relationship of the injection correction amount to the injection correction amount of the stable cylinder,
The air-fuel ratio feedback control is performed by reflecting the injection correction amount of the stable cylinder in the fuel injection amount of the stable cylinder and reflecting the injection correction amount of the unstable cylinder in the fuel injection amount of the unstable cylinder. It is characterized by performing.

The twelfth invention is the eleventh invention, in which
The control unit is
If the determination that the pre-stabilization change of the thermal strain characteristic has been settled is denied for all the cylinders, then immediately grasp the air-fuel ratio features of all the cylinders,
When a plurality of cylinders are determined to be the stable cylinders, the injection correction amount of the cylinder that is determined to be the stable cylinder earliest is used as the basis of the injection correction amount of the unstable cylinder.

  According to the first aspect of the invention, the degree to which the output of the in-cylinder pressure sensor is reflected in the control of the actuator can be suppressed until the pre-stabilization change of the thermal strain characteristic of the in-cylinder pressure sensor is settled. For this reason, according to the present invention, when the output characteristic of the in-cylinder pressure sensor changes due to the pre-stability change of the thermal strain characteristic, the error superimposed on the sensor output is directly reflected in the actuator control of the internal combustion engine. Can be avoided.

  According to the second invention, the mass combustion ratio MFBax at the specific crank angle ax belonging to the region from the maximum in-cylinder pressure generating crank angle to the exhaust valve opening crank angle is calculated. The influence of the thermal strain of the pressure sensitive part tends to appear in the output of the in-cylinder pressure sensor after the maximum in-cylinder pressure is generated and before the exhaust valve is opened. The mass combustion ratio appears in a state where the output characteristic of the in-cylinder pressure sensor is amplified. For this reason, whether or not the mass combustion ratio MFBax at the specific crank angle ax has reached a stable value indicates that the output characteristic of the in-cylinder pressure sensor is stable, that is, the pre-stabilization change of the thermal strain characteristic of the pressure-sensitive part. It is possible to accurately determine whether it has been settled. Therefore, according to the present invention, it is possible to correctly determine the convergence of the pre-stability change of the thermal strain characteristics.

  According to the third aspect of the present invention, it is determined whether or not the pre-stabilization change of the thermal strain characteristic is settled based on the output Pcax of the in-cylinder pressure sensor at the specific crank angle ax. The specific crank angle ax belongs to a crank angle region in which the influence of thermal distortion of the pressure-sensitive portion is likely to appear in the output of the in-cylinder pressure sensor. For this reason, according to the present invention, it is possible to appropriately determine the convergence of the pre-stable change of the thermal strain characteristics.

  According to the fourth invention, the specific crank angle ax is set to the crank angle immediately before the exhaust valve is opened. The degree of influence of the thermal strain of the pressure-sensitive part on the sensor output increases as the in-cylinder pressure decreases due to the S / N ratio. For this reason, according to the present invention, it is possible to determine whether or not the pre-stabilization change of the thermal strain characteristic is settled at the timing when the influence of the thermal strain is sufficiently reflected in the output of the in-cylinder pressure sensor.

  According to the fifth invention, when the mass combustion ratio MFBax of the specific crank angle ax is sufficiently close to the reference mass combustion ratio MFBax0 which is the original value, the pre-stabilization change of the thermal strain characteristic is settled. Judgment can be made.

  According to the sixth invention, when the difference between the mass combustion rate MFBax1 and the mass combustion rate MFBax is sufficiently small, it is determined that the mass combustion rate MFBax has reached a stable value. Both are mass combustion ratios calculated continuously in time under conditions that can be considered to generate the same mass combustion ratio. If the difference between the two becomes sufficiently small, it can be determined that the mass combustion ratio MFBax has converged to a stable value, and therefore, it can be determined that the pre-stabilization change of the thermal strain characteristics has settled.

  According to the seventh invention, when the fuel cut is switched on and off, the drift amount appearing in the output of the in-cylinder pressure sensor is calculated. This drift amount increases as the deposit builds up on the in-cylinder pressure sensor. On the other hand, the pre-stabilization change in thermal strain characteristics tends to converge as the deposit builds up. For this reason, the drift amount of the in-cylinder pressure sensor can be used as a characteristic value for determining the degree of convergence of the change before stabilization. Therefore, according to the present invention, it is possible to appropriately determine whether or not the pre-stabilization change of the thermal strain characteristic has been settled based on the drift amount.

  According to the eighth aspect, until the thermal strain characteristic of the in-cylinder pressure sensor is stabilized, the state determination based on the output is prohibited. For this reason, according to the present invention, it is possible to prevent the command value calculation method for the actuator from being erroneously switched due to the influence of the sensor error due to the thermal strain characteristics before stabilization.

  According to the ninth aspect of the invention, the feedback gain when reflecting the output of the in-cylinder pressure sensor in the command value of the actuator can be lowered until the thermal strain characteristic of the in-cylinder pressure sensor is stabilized. For this reason, according to this invention, it can prevent that the influence of the sensor error resulting from the thermal-strain characteristic before stabilization is fed back excessively with respect to an actuator.

  According to the tenth aspect, the feedback control of the actuator based on the output of the in-cylinder pressure sensor can be prohibited until the thermal strain characteristic of the in-cylinder pressure sensor is stabilized. For this reason, according to this invention, it can prevent that the influence of the sensor error resulting from the thermal-strain characteristic before stabilization is fed back to an actuator.

  According to the eleventh aspect, the air-fuel ratio feature of each cylinder is grasped based on the output of the in-cylinder pressure sensor before the thermal strain characteristic is stabilized (hereinafter referred to as “pre-stability output”). The effect of thermal strain appears significantly in the output before stabilization. On the other hand, the influence of thermal strain on the sensor output increases as the temperature in the cylinder in which the sensor is mounted increases. The in-cylinder temperature has a correlation with the air-fuel ratio of the air-fuel mixture combusted in the cylinder, and becomes higher as the air-fuel ratio is richer. Therefore, the output before stabilization is a value that greatly includes the influence of thermal distortion in a cylinder in which the air-fuel ratio tends to be rich, that is, in a cylinder with a rich feature. On the other hand, in a cylinder with a lean feature, the value does not include the effect of thermal strain. For this reason, the output before stabilization of each cylinder shows a correlation with the air-fuel ratio feature of each cylinder. Therefore, in the present invention, the air-fuel ratio feature of each cylinder can be grasped based on the pre-stabilization output.

  If the air-fuel ratio feature of each cylinder can be grasped, the relative relationship of the amount of fuel to be injected into each cylinder can be specified in order to make the air-fuel mixture of all the cylinders have the same air-fuel ratio. When the injection correction amount to be fed back is determined for at least one cylinder, the injection correction amount to be fed back can be determined for the other cylinders by applying the above relative relationship to the injection correction amount.

  In the present invention, for the stable cylinder, an appropriate injection correction amount is calculated based on the output of the in-cylinder pressure sensor in which the thermal strain characteristics have already fallen. For an unstable cylinder, the injection correction amount can be calculated without depending on the sensor output of the cylinder by applying the relative relationship of the air-fuel ratio feature to the injection correction amount of the stable cylinder. For this reason, according to the present invention, appropriate air-fuel ratio feedback control can be started for all the cylinders when the stability of the sensor characteristics for one cylinder has been confirmed.

  According to the twelfth invention, when the internal combustion engine is new, the determination that the pre-stabilization change of the thermal strain characteristic has been settled is denied in all the cylinders. In this case, the air-fuel ratio features of all the cylinders are immediately grasped thereafter. Similarly, after the deposits of all the cylinders are removed by cleaning or the like, the air-fuel ratio features of all the cylinders are immediately grasped thereafter. As described above, according to the present invention, it is possible to appropriately grasp the air-fuel ratio features of all the cylinders in a situation where the pre-stabilization output is obtained in all the cylinders. Further, according to the present invention, when the plurality of cylinders are stable cylinders, the injection correction amount of the unstable cylinder is calculated based on the cylinder that is determined to be the stable cylinder earliest. Therefore, according to the present invention, in a transition period in which a plurality of unstable cylinders sequentially become stable cylinders, or in a situation where only some of the cylinders become unstable cylinders due to dropout of deposits, air-fuel ratio feedback of all cylinders Control can be continued without confusion.

It is a figure which shows the structure of Embodiment 1 of this invention. It is a figure for demonstrating the influence of the thermal strain superimposed on the output of a cylinder pressure sensor. It is a figure showing a mode that the mass combustion rate MFB calculated based on the output of a cylinder pressure sensor converges with stabilization of a thermal strain characteristic. It is a figure showing a mode that the mass combustion ratio MFBax in the specific crank angle ax converges to a stable value with the increase in the operation time of an internal combustion engine. It is a flowchart of an example of the routine which a control apparatus performs in Embodiment 1 of this invention. It is a flowchart of the modification of the routine which a control apparatus performs in Embodiment 1 of this invention. It is a figure showing a mode that MFB increases with progress of combustion after ignition timing SA. It is a figure which shows the relationship between the air fuel ratio (A / F) of air-fuel | gaseous mixture, and the period from ignition timing SA to crank angle CA10 from which MFB becomes 10%. It is a figure showing a mode that MFBax in each of a plurality of cylinders converges to a stable value sequentially with an increase in operating time of an internal-combustion engine. It is a figure showing the relationship between the MFB curve calculated based on the pre-stabilization output of a cylinder pressure sensor, and the air-fuel ratio feature of a cylinder. It is a figure showing the relationship between the MFBax computed based on the pre-stabilization output of a cylinder pressure sensor, and the air fuel ratio feature of a cylinder. It is a figure showing the relationship between the MFBax calculated based on the pre-stabilization output of a cylinder pressure sensor, and the air fuel ratio of air-fuel mixture. It is a figure which shows the relative relationship of the air fuel ratio feature of each cylinder. It is a figure which shows the relative relationship of the injection correction amount which should be applied to each cylinder in order to control all the cylinders to the same air fuel ratio. It is a figure for demonstrating the method of calculating the injection correction amount of an unstable cylinder ((triangle | delta), (square), (circle)) based on the injection correction amount of a stable cylinder (*). It is a figure showing a mode that an air fuel ratio feedback is started with respect to all the cylinders by making the first stable cylinder into a representative cylinder. It is a flowchart (the 1) of an example of the routine which a control apparatus performs in Embodiment 2 of this invention. It is a flowchart (the 2) of an example of the routine which a control apparatus performs in Embodiment 2 of this invention. It is a figure which shows the mode of the drift which appears in the output of a cylinder internal pressure sensor at the time of the return from a fuel cut. It is the figure which rewrote the state shown in FIG. 19 paying attention to drift amount. It is a figure showing the calculation method of the relationship between reference | standard drift amount (DELTA) P (alpha) and the driving | running state of an internal combustion engine, and drift reduction amount (DELTA) P (alpha)-(DELTA) P1. It is a figure which shows the relationship between the amount of drift reduction, and the deposit adhesion amount. It is a flowchart of an example of the routine which a control apparatus performs in Embodiment 3 of this invention.

  Embodiments of the present invention will be described below with reference to the drawings. In addition, in each figure, the same code | symbol is attached | subjected to the element which is common or respond | corresponds, and the overlapping description is abbreviate | omitted. The present invention is not limited to the following embodiments.

Embodiment 1 FIG.
[Configuration of Embodiment 1]
FIG. 1 shows a configuration of an internal combustion engine 10 according to Embodiment 1 of the present invention. The internal combustion engine 10 includes a piston 14 in a cylinder 12. A spark plug 16 is provided in the upper center of the cylinder 12. A fuel injection valve 18 is disposed in the vicinity of the spark plug 16.

  An in-cylinder pressure sensor (CPS, Combustion Pressure Sensor) 20 is further disposed on the upper portion of the cylinder 12. The in-cylinder pressure sensor 20 is a sensor that generates an output corresponding to the in-cylinder pressure Pc, and has a pressure-sensitive part 22 exposed in the cylinder. More specifically, the in-cylinder pressure sensor 20 includes a diaphragm and a shaft member disposed inside the pressure-sensitive portion 22. When the pressure in the cylinder 12 fluctuates, the fluctuation is transmitted to the inside of the sensor by the diaphragm and the shaft member of the pressure sensing unit 22. The in-cylinder pressure sensor 20 generates an output corresponding to the in-cylinder pressure Pc by converting the propagating force into an electrical signal.

  An intake passage 24 communicates with the cylinder 12. An intake valve 26 is disposed at a port portion where the intake passage 24 communicates with the cylinder 12. A throttle valve 28 is disposed upstream of the intake valve 26. The throttle valve 28 is an electronically controlled valve mechanism that operates according to the accelerator opening, and the intake air amount can be changed according to the opening.

  An exhaust passage 30 communicates with the cylinder 12 of the internal combustion engine 10. An exhaust valve 32 is incorporated in the exhaust passage 30 at a portion communicating with the cylinder 12. The internal combustion engine 10 incorporates various sensors generally mounted on the internal combustion engine, such as a crank angle sensor 34 mounted on the crankcase and a water temperature sensor 36 mounted on the cylinder block.

  As shown in FIG. 1, the system according to the present embodiment includes an electronic control unit (ECU) 40. The ECU 40 is electrically connected to the above-described actuators such as the ignition plug 16 and the fuel injection valve 18 and sensors such as the crank angle sensor 34 and the water temperature sensor 36.

[Thermal strain characteristics of in-cylinder pressure sensor 20]
FIG. 2 is a diagram for explaining the influence of thermal strain superimposed on the output of the in-cylinder pressure sensor 20. In FIG. 2, the line denoted by reference numeral 42 indicates the true value Pc0 of the in-cylinder pressure. The true value Pc0 can be detected by, for example, arranging a measuring instrument with a much higher accuracy than the in-cylinder pressure sensor 20 in the cylinder 12 and performing measurement.

  A line indicated by reference numeral 44 in FIG. 2 indicates the in-cylinder pressure Pc calculated by the ECU 40 based on the output of the in-cylinder pressure sensor 20. The pressure sensitive part 22 of the in-cylinder pressure sensor 20 is exposed to heat in the cylinder 12. If the pressure sensitive part 22 is exposed to high temperature heat, thermal distortion will occur in the diaphragm arranged there. If thermal distortion occurs in the diaphragm, the stress due to the distortion is transmitted to the shaft member, and the output of the in-cylinder pressure sensor 20 becomes a value deviated from the true value Pc0 of the in-cylinder pressure. Whether the sensor output shifts to the plus side or the minus side with respect to the true value Pc0 depends on the structure of the sensor, but in the in-cylinder pressure sensor 20 of the present embodiment, the shift occurs on the minus side. Therefore, the in-cylinder pressure Pc based on the sensor output becomes a value shifted in a smaller direction with respect to the true value Pc0, as shown in FIG.

  In the pressure sensitive part 22, the greater the amount of heat received, the greater the thermal strain. The amount of heat received by the pressure-sensitive portion 22 increases as the explosion process proceeds. For this reason, a large error is unlikely to occur in the sensor output during the first half of the explosion stroke with a small amount of heat. Further, under the condition where the true value Pc0 of the in-cylinder pressure is large, the influence of thermal strain becomes relatively small, and therefore, the in-cylinder pressure Pc based on the sensor output is unlikely to deviate from the true value Pc0 due to the S / N ratio. For this reason, the in-cylinder pressure Pc based on the sensor output is substantially equal to the true value Pc0 during the first half of the explosion stroke, that is, until the maximum in-cylinder pressure is generated from the ignition timing.

  On the other hand, in the latter half of the explosion stroke where the amount of heat received by the pressure sensing unit 22 increases and the true value Pc0 of the in-cylinder pressure decreases, the output error due to thermal strain becomes significant, and the in-cylinder pressure Pc based on the sensor output is reduced. The value tends to deviate from the true value Pc0. The deviation becomes larger as the explosion stroke proceeds, and usually becomes the largest immediately before the exhaust valve 32 is opened.

[Relationship between deposit amount and thermal strain]
Deposits generated in the cylinder 12 as the air-fuel mixture burns adhere to the pressure-sensitive portion 22. The deposited amount of deposit increases as the operating time of the internal combustion engine 10 increases. The deposit has the effect of hindering the propagation of heat. For this reason, the error due to thermal strain decreases as the deposit builds up, and eventually converges to a stable value. That is, in the output of the in-cylinder pressure sensor 20, an error due to thermal strain appears significantly at the stage where no deposit is attached. This error is greatly reduced at the initial stage where deposits are accumulated in the pressure-sensitive portion 22, and eventually converges to a stable value as the operating time of the internal combustion engine 10 elapses.

  In the case of the in-cylinder pressure sensor 20 used in the present embodiment, specifically, the pre-stabilization change of the thermal strain characteristic is settled around the deposit accumulation amount reaching 100 μm, and the output error caused by the thermal strain converges to the stable value. To do. When converted to the operation time of the internal combustion engine 10, the change before thermal stabilization of the in-cylinder pressure sensor 20 converges in about two days. Hereinafter, the sensor output before the change before the stabilization of the thermal strain characteristic is settled is referred to as “pre-stability output”, and the sensor output after the change is settled is referred to as “post-stabilization output”.

FIG. 3 is a diagram showing how the mass combustion ratio MFB calculated based on the output of the in-cylinder pressure sensor converges with stabilization of the thermal strain characteristics. The mass combustion ratio MFB is the mass ratio of the fuel burned in the cylinder 12 during the explosion stroke. The mass combustion ratio MFB _{θ at the} crank angle θ can be calculated by the following equation.
MFB = {Pc (θ) ・ V (θ) ^κ -Pc (θs) ・ V (θs) ^κ } / {Pc (θe) ・ V (θe) ^κ -Pc (θs) ・ V (θs) ^κ }
... (1 set)
Where V (θx) is the volume in the cylinder 12 at the crank angle θx, Pc (θx) · V (θx) ^κ is the amount of heat generated until the crank angle reaches θx, κ is the specific heat ratio of the mixture, and θs is the combustion The crank angle regarded as the start and θe is the crank angle regarded as the end of combustion.

  The characteristic diagram shown on the left side of FIG. 3 is a diagram in which the MFB data 46 based on the output before stabilization is superimposed on the MFB data 46 based on the output of the measuring instrument (true value Pc0 of the in-cylinder pressure). More specifically, the MFB data 48 is a set of MFBs calculated based on the output of the new in-cylinder pressure sensor 20. Each of the MFB data 46 and 48 includes a large number of MFB results obtained in a process in which the explosion stroke is repeated many times.

As shown in the above characteristic diagram, the MFB data 46 based on the true value Pc0 has a stable waveform with little variation. On the other hand, the MFB data 48 based on the pre-stabilization output is lower than the MFB data 46 in the second half of the explosion stroke, and has a considerable variation. The reason why the MFB data 48 before stabilization shows such characteristics is as follows.
1. The output of the in-cylinder pressure sensor 20 becomes a value smaller than the true value Pc0 due to the influence of thermal strain in the second half of the explosion stroke.
2. No deposit is attached to the new cylinder pressure sensor 20. In the process of deposit accumulation, the output characteristics of the in-cylinder pressure sensor 20 change abruptly. As a result, the MFB calculated for each explosion stroke also varies greatly.

  The characteristic diagram shown on the right side of FIG. 3 is a diagram in which the MFB data 46 based on the output after stabilization is superimposed on the MFB data 46 based on the output of the measuring instrument (true value Pc0 of the in-cylinder pressure). As described above, the output characteristics of the in-cylinder pressure sensor 20 are stabilized when the deposit adhesion amount reaches about 100 μm. Then, according to the post-stabilized output generated from the in-cylinder pressure sensor 20 after stabilization, the MFB data 50 roughly matching the MFB data 46 based on the true value Pc0 can be calculated as shown in the figure.

[Features of Embodiment 1]
FIG. 4 shows a process until the mass combustion ratio MFBax reaches a stable value at a specific crank angle ax. Time 0 in FIG. 4 indicates a time point when the in-cylinder pressure sensor 20 is new, that is, a time point when no deposit is attached to the in-cylinder pressure sensor 20. Further, the specific crank angle ax is a crank angle belonging to a region from the crank angle at which the maximum in-cylinder pressure is generated to the crank angle at which the exhaust valve 32 is opened, that is, the output of the in-cylinder pressure sensor 20. This is a crank angle at which the influence of thermal strain tends to appear remarkably. Here, the crank angle at which the mass combustion ratio MFB reaches almost 100% is ax, and specifically, 90 ° CA (ATDC 90 ° CA) after compression top dead center is ax.

  The mass combustion rate MFBax of the specific crank angle ax is calculated to an excessive value at time 0 due to the influence of thermal distortion, and thereafter increases with the passage of operating time, and eventually converges to a stable value near 100%. If the MFBax changes in this way as the deposit increases, an appropriate reference value 52 is provided so that sufficient deposit adheres to the pressure-sensitive portion 22 or is subtracted. It is possible to appropriately determine whether or not the pre-stabilization change of the thermal strain characteristics has converged. If it is possible to appropriately determine whether or not the pre-stabilization change of the thermal strain characteristic has been settled, the output of the in-cylinder pressure sensor 20 and further the mass combustion ratio MFB calculated based on the output are reliable values. It can be correctly determined whether or not.

  In the present embodiment, the mass combustion ratio MFB is used for ignition timing control in which the ignition plug 16 is controlled, air-fuel ratio feedback control in which the fuel injection valve 18 is controlled, and the like. The output of the in-cylinder pressure sensor 20 is used for state determination such as torque fluctuation determination and misfire determination (the result is reflected in, for example, switching of a command value calculation method for the spark plug 16 and the fuel injection valve 18). ) While the mass combustion ratio MFB and the output of the in-cylinder pressure sensor 20 are unreliable, it is desirable to avoid inappropriate reflection of them in the operating state of the internal combustion engine 10. Therefore, in the present embodiment, the degree to which the output of the in-cylinder pressure sensor 20 and the mass combustion ratio MFB are reflected in the control of the spark plug 16 and the fuel injection valve 18 until it is determined that the MFBax has converged to a stable value. Decided to lower. More specifically, until it is determined that the MFBax has converged to a stable value, the control based on the output of the in-cylinder pressure sensor 20 and the mass combustion ratio MFB is prohibited, and when the determination is affirmed, It was decided to allow control based on the above.

[Operation of Embodiment 1]
FIG. 5 is a flowchart of an example of a routine executed by the ECU 40 in the present embodiment. The routine shown in FIG. 5 is executed every cycle for each cylinder of the internal combustion engine 10. When the routine shown in FIG. 5 is started, first, the mass combustion ratio MFB is calculated (step 100). Specifically, the in-cylinder pressure Pc and the volume V are read every predetermined sampling time, and the MFB for one cycle is calculated by sequentially applying them to the above (formula 1). The ECU 40 stores a map that defines the volume V in relation to the crank angle. Here, the volume V is specified according to the map.

  Next, the mass combustion ratio MFBax at the specific crank angle ax is read out from the calculated MFB (step 102).

Next, a reference mass combustion ratio MFBax0 is calculated (step 104). The reference mass combustion rate MFBax0 is the mass combustion rate that should originally occur at the specific crank angle ax, in other words, the mass combustion rate obtained at the specific crank angle ax when the calculation is performed based on the true value Pc0. . The reference mass combustion ratio MFBax0 is a value uniquely determined with respect to the operating state of the internal combustion engine 10. In the present embodiment, the ECU 40 stores a map that defines MFBax0 in relation to the engine speed NE and the engine load KL. In step 104, specifically, MFBax0 is calculated by the following processing.
1. Identification of current NE and KL (NE and KL that generated MFB calculated in step 100).
2. Reading out the above map based on the identified NE and KL.

When the above processing is completed, it is next determined whether or not the mass combustion rate MFBax calculated in the current routine is sufficiently close to the reference mass combustion rate MFBax0. Specifically, it is determined whether or not the relationship of the following equation is established (step 106).
MFBax0−MFBax ≦ judgment value (2 formulas)

  The output of the in-cylinder pressure sensor 20 becomes a value that matches the true value Pc0 of the in-cylinder pressure as the deposit amount increases and the pre-stabilization change of the thermal strain characteristic is settled. Therefore, the calculated value of the mass combustion ratio MFBax approaches the reference mass combustion ratio MFBax0 as the pre-stabilization change is settled. For this reason, if the difference between the two becomes sufficiently small, it can be determined from the fact that the pre-stabilization change of the thermal strain characteristics has been settled. The “determination value” in the above (Equation 2) is a value set so that this determination is possible. Therefore, if the determination in step 106 is affirmative, it can be determined that the pre-stabilization change in the thermal strain characteristics of the pressure-sensitive portion 22 has subsided.

  If it is determined in step 106 that the relationship of (Expression 2) is not established, the current routine is terminated. In the initial state, the ECU 40 is configured to prohibit the control based on the output of the in-cylinder pressure sensor 20 and the mass combustion ratio MFB. Until it is determined that (Formula 2) is established, the initial state is maintained, and execution of control based on the output of the in-cylinder pressure sensor 20 is prohibited. For this reason, according to the present embodiment, it is possible to prevent the output of the in-cylinder pressure sensor 20 from being improperly reflected in the operating state of the internal combustion engine 10 under a situation in which the thermal strain characteristic is greatly changed. .

  On the other hand, if the establishment of (Formula 2) is recognized in Step 106, it is determined that the pre-stabilization change of the thermal strain characteristics has been settled (Step 108). That is, it is determined that the change in sensor error due to thermal strain is stable.

  Even if an error is superimposed on the sensor output, if the error is stable, the internal combustion engine 10 can be appropriately controlled based on the sensor output. For this reason, the ECU 40 permits the execution of the control based on the output of the in-cylinder pressure sensor 20 following the step 108 (step 110). Thereafter, the ECU 40 performs ignition timing control, air-fuel ratio feedback control based on the mass combustion ratio MFB, torque fluctuation determination and misfire determination based on the output of the in-cylinder pressure sensor 20. According to the above processing, the operating state of the internal combustion engine 10 can be appropriately controlled based on the output of the in-cylinder pressure sensor 20 with stable thermal strain characteristics.

[Modification of Embodiment 1]
By the way, in the first embodiment described above, it is determined that the pre-stabilization change of the thermal strain characteristic has been settled when the mass combustion rate MFBax is sufficiently close to the reference mass combustion rate MFBax0 (see step 106 above). However, the method for determining the convergence of the pre-stability change of the thermal strain characteristics is not limited to this.

  FIG. 6 is a flowchart for explaining another example of determining the convergence of the pre-stability change of the thermal strain characteristic. In this flowchart, at the end of the routine, the mass combustion ratio MFBax calculated in the current routine is stored as the previous value MFBax1 (step 116). MFBax is a value that should be uniquely determined with respect to the engine speed NE and the engine load KL. More specifically, the current mass combustion ratio MFBax is stored as the previous value MFBax1 in association with (NE, KL) at the time of execution of the current routine.

In the routine shown in FIG. 6, following the steps 100 and 102, the previous value MFBax1 is compared with the mass combustion ratio MFBax calculated this time (step 112). Here, specifically, the following processing is performed.
1. Identification of current (NE, KL) (which generated MFB in step 100 (NE, KL)).
2. Reading the previous value MFBax1 stored in relation to (NE, KL) and the mass combustion ratio MFBax calculated this time.

Next, it is determined whether or not the difference between the previous value MFBax1 read in step 112 and the current value MFBax is sufficiently small. Specifically, it is determined whether or not the relationship of the following equation is established (step 114).
| MFBax1−MFBax | ≦ judgment value (3 formulas)

  As described with reference to FIG. 4, the mass combustion rate MFBax of the specific crank angle ax converges to a stable value (approximately 100%) as the operation time elapses. At this time, the difference between the previous value MFBax1 and the current value MFBax decreases as the convergence of the MFBax progresses. Therefore, if the difference between the previous value MFBax1 and the current value MFBax calculated under the same (NE, KL) is sufficiently small, it can be determined that the mass combustion ratio MFBax has converged to a stable value. The “determination value” in the above (Expression 3) is a value set so that this determination is possible. Therefore, if the determination in step 114 is affirmed, it can be determined that the pre-stabilization change in the thermal strain characteristics of the pressure-sensitive portion 22 has been settled, as in the case where the determination in step 106 shown in FIG. 5 is affirmed. it can.

  In Embodiment 1 described above, control of the internal combustion engine 10 based on the output is prohibited until the characteristics of the in-cylinder pressure sensor 20 are stabilized, but the present invention is not limited to this. . For example, in the present embodiment, feedback control of ignition timing and air-fuel ratio is performed based on the mass combustion ratio MFB. For these, the control may be performed by lowering the feedback gain instead of prohibiting the control until the pre-stabilization change of the thermal strain characteristic is settled. In this embodiment, the occurrence of torque fluctuation or misfire is determined based on the output of the in-cylinder pressure sensor 20. For these, the determination may not be prohibited until the pre-stability change of the thermal strain characteristic is settled, but the possibility of erroneous determination may be reduced by tightening the determination conditions. Furthermore, the execution of some controls may be prohibited, the gain may be reduced for the remaining controls, or the judgment conditions may be stricter.

  Further, in the first embodiment described above, it is determined based on the mass combustion ratio MFBax whether or not the pre-stabilization change of the thermal strain characteristic of the pressure-sensitive part 22 is settled. In the mass combustion ratio MFBax, as shown in the above (formula 1), the gain resulting from the volume V of the cylinder 12 and the specific heat ratio κ is added to the in-cylinder pressure Pc. Therefore, the mass combustion ratio MFBax appears in a state where the influence of the output error is emphasized as compared with the output itself of the in-cylinder pressure sensor 20. In this regard, the mass combustion ratio MFBax is a suitable parameter for determining the convergence of the pre-stabilization change of the thermal strain characteristics. However, the output of the in-cylinder pressure sensor 20 is also affected by the thermal strain characteristics of the pressure-sensitive portion 22, and the above determination can be made from the sensor output in the latter half of the explosion stroke. Specifically, when the sensor output Pcax at the specific crank angle ax reaches a determination value for determining the convergence of the pre-stabilization change, it may be determined that the pre-stabilization change of the thermal strain characteristic has subsided. Alternatively, when the difference between the previous value and the current value of the sensor output Pcax at a specific crank angle ax becomes smaller than the judgment value for judging the convergence of the pre-stability change, the pre-stabilization change of the thermal strain characteristics is reduced. It may be determined that

  In the first embodiment described above, the specific crank angle ax is set to ATDC 90 ° CA, but the value is not limited to this. As described with reference to FIG. 2, the output error of the in-cylinder pressure sensor 20 due to thermal strain becomes more prominent as the explosion stroke proceeds. For this reason, the specific crank angle ax is preferably closer to the valve opening timing of the exhaust valve 32 in determining the convergence of the pre-stabilization change of the thermal strain characteristic. Therefore, the specific crank angle ax may be set anywhere in the region from the generation timing of the maximum in-cylinder pressure to the valve opening timing of the exhaust valve 32, and in particular, it should be set in the second half of the region. Is desirable. Furthermore, it is desirable to set the crank angle immediately before the exhaust valve 32 is opened (the crank angle corresponding to the last sampling timing before the exhaust valve 32 is opened).

  Further, in the first embodiment described above, a situation where the pre-stabilization change of the thermal strain characteristic occurs is exemplified as a situation where the in-cylinder pressure sensor 20 is new, and the degree to which the sensor output is reflected in the control under the situation. We are going to lower it. However, the situation where the degree to which the sensor output is reflected in the control is not limited to this. That is, the thermal strain characteristic of the pressure-sensitive portion 22 shows a large change before stabilization under the condition that the deposit amount is small. The deposit amount is not limited to when the in-cylinder pressure sensor 20 is new, but also immediately after the pressure-sensitive portion 22 is cleaned or immediately after the deposit is dropped from the pressure-sensitive portion 22 due to the influence of combustion pressure or the like. Arise. For this reason, after the control based on the in-cylinder pressure sensor 20 is once permitted by the processing in step 110, if the determination in step 106 or step 114 is denied again, the in-cylinder pressure sensor is used until the determination is affirmed again. The degree at which the output of 20 is reflected in the control may be lowered again.

Embodiment 2. FIG.
Next, a second embodiment of the present invention will be described with reference to FIGS. The system of the present embodiment can be realized using the hardware shown in FIG. However, in the present embodiment, the internal combustion engine 10 is a four-cylinder engine.

[Air-fuel ratio feedback control based on MFB]
In the present embodiment, the ECU 40 performs air-fuel ratio feedback control based on the mass combustion ratio MFB, as in the case of the first embodiment. Since the characteristic operation of the present embodiment is closely related to the contents of the air-fuel ratio feedback control, the outline of the control will be described below.

  FIG. 7 shows a change in the mass combustion ratio MFB accompanying a change in the crank angle. In FIG. 7, “SA” represents the ignition timing. “CA10” represents a crank angle at which the mass combustion ratio MFB is 10%. The ECU 40 calculates SA on the premise that an ignition command is issued to the spark plug 16. Therefore, SA is a parameter that can be read by the ECU 40. The ECU 40 can calculate the MFB based on the above (formula 1), and can read the crank angle at which the MFB is 10% as CA10.

  FIG. 8 shows the relationship between the air-fuel ratio (A / F) of the air-fuel mixture and SA-CA10. SA-CA10 is the width of the crank angle from the ignition timing to CA10. SA-CA10 has a smaller value as the combustion of the air-fuel mixture progresses faster after the ignition timing. The combustion of the air-fuel mixture proceeds faster as the air-fuel ratio is richer. For this reason, as shown in FIG. 8, SA-CA10 has a smaller value as the air-fuel ratio is smaller (richer), and becomes a larger value as the air-fuel ratio becomes larger (lean).

  The relationship shown in FIG. 8 can be grasped in advance for each cylinder 12 of the internal combustion engine 10. If this relationship is known, it is possible to detect what air-fuel ratio the air-fuel mixture has based on SA-CA10.

  In the present embodiment, the ECU 40 can set the target value of SA-CA10 corresponding to the target air-fuel ratio. If the actual SA-CA10 is smaller than the target value, it is determined that the air-fuel ratio of the air-fuel mixture is smaller (rich) than the target air-fuel ratio, and the fuel injection amount is reduced. On the other hand, when the actual SA-CA10 is larger than the target value, it is determined that the air-fuel ratio of the air-fuel mixture is larger than the target air-fuel ratio (lean), and the fuel injection amount is increased. The ECU 40 thus performs air-fuel ratio feedback control based on the mass combustion ratio MFB.

[Influence of variation between cylinders]
FIG. 9 shows how the mass combustion ratios MFBax of the four cylinders 12 converge toward a stable value independently after the new internal combustion engine 10 starts to operate. In FIG. 9, the line denoted by reference numeral 52 is a reference value for determining the stable value convergence of the MFBax, as in the case of FIG.

  The pre-stabilization change of the thermal strain characteristic of the in-cylinder pressure sensor 20 tends to converge as deposit deposition proceeds. On the other hand, the deposit accumulation speed of each cylinder 12 varies depending on the condition of each cylinder 12.

  For example, the air-fuel ratio of each cylinder 12 is not necessarily the same for all cylinders due to individual differences in the fuel injection valves 18 and cylinder differences in the intake system. That is, even if the same command is issued uniformly to the fuel injection valves 18 of all the cylinders 12, the air-fuel ratio is biased to the rich side for some cylinders, while the air-fuel ratio is lean for some cylinders. The situation of being biased to the side can occur as a normal matter. In the cylinder 12 having a rich air-fuel ratio feature, a larger amount of deposit is easily generated than in the cylinder 12 having a lean feature. The deposit accumulation rate is generated due to, for example, such a difference in air-fuel ratio features between cylinders.

  FIG. 9 illustrates, as a general phenomenon, how MFBax converges to a stable value at an earlier stage in a cylinder that has a richer air-fuel ratio feature and a larger amount of deposit is likely to be generated. However, in addition to the air-fuel ratio feature, there are various factor differences between the cylinders that make the time required for MFBax stabilization different. For this reason, the difference in time required for the stabilization of MFBax does not necessarily mean a difference in air-fuel ratio characteristics between cylinders.

  In the first embodiment described above, the air-fuel ratio feedback control based on the output of the in-cylinder pressure sensor 20 is started after the mass combustion ratio MFBax has converged to a stable value. When this method is applied to each of the four cylinders, the air-fuel ratio feedback control is started in all the cylinders 12 when the convergence of the MFBax is determined in the cylinder having the slowest convergence (in FIG. Point in time). As described above, when the method of Embodiment 1 is simply applied to a multi-cylinder internal combustion engine, air-fuel ratio feedback based on the mass combustion ratio MFB is applied to a cylinder whose MFBax converges slowly due to variations among cylinders. A situation may arise where control cannot be initiated over a long period of time.

[Features of Embodiment 2]
In the system of the present embodiment, a cylinder in which convergence of MFBax is recognized (hereinafter referred to as “stable cylinder”) and a cylinder in which the convergence is not recognized (hereinafter referred to as “unstable cylinder”) are mixed. Further, the present invention is characterized in that air-fuel ratio feedback control for the unstable cylinder is executed based on the output of the cylinder pressure sensor 20 for the stable cylinder. Hereinafter, this feature point will be described with reference to FIGS. 10 to 16.

(Air-fuel ratio feature of cylinder)
FIG. 10 is a diagram for explaining the relationship between MFBax and the air-fuel ratio feature of each cylinder. Each of the three waveforms shown in FIG. 10 shows the calculation result of the mass combustion ratio MFB based on the pre-stabilization output of the in-cylinder pressure sensor 20, that is, the sensor output under the condition that the deposit hardly adheres. Among these, the central waveform shows the MFB result when the air-fuel ratio of the air-fuel mixture is a standard value. On the other hand, the left waveform shows the MFB result when the air-fuel ratio of the air-fuel mixture is shifted to the rich side. On the other hand, the right waveform shows the MFB result when the air-fuel ratio of the air-fuel mixture is shifted to the lean side.

  The pressure sensitive part 22 of the in-cylinder pressure sensor 20 generates a larger thermal strain as the temperature in the cylinder 12 becomes higher. In particular, under a situation where the deposit amount is small, the temperature difference in the cylinder 12 causes a significant difference in the magnitude of thermal strain. The mass combustion ratio MFBax of the specific crank angle ax is closer to 100% as the thermal strain is smaller, and is smaller as the thermal strain is larger. For this reason, a difference in temperature in the cylinder 12 appears remarkably in the mass combustion ratio MFBax of the specific crank angle ax based on the pre-stabilization output of the in-cylinder pressure sensor 20.

  The temperature in the cylinder 12 has a correlation with the air-fuel ratio of the air-fuel mixture. Specifically, the temperature in the cylinder 12 becomes higher as the combustion air-fuel mixture becomes richer, and becomes lower as the air-fuel ratio becomes leaner. For this reason, as shown in FIG. 10, the mass combustion ratio MFBax of the specific crank angle ax becomes larger as the air-fuel mixture becomes leaner, while it becomes smaller as the air-fuel mixture becomes richer.

  FIG. 11 is a diagram in which the above relationships shown in FIG. FIG. 12 is a diagram showing the relationship shown in FIG. 11 with the vertical axis and the horizontal axis interchanged for convenience of explanation. The relationship shown in FIG. 12, that is, the relationship between the MFBax based on the pre-stabilization output and the air-fuel ratio, is determined for the hardware configuration if the operating state of the internal combustion engine 10 is fixed. Therefore, this relationship can be grasped for the internal combustion engine 10.

  FIG. 13 is a diagram in which the air-fuel ratio features of four cylinders (◯, □, ☆, Δ) are superimposed on the characteristic diagram shown in FIG. In the present embodiment, the ECU 40 calculates the mass combustion ratio MFBax for each cylinder in a situation where the in-cylinder pressure sensors 20 of all the cylinders 12 emit pre-stabilization output (for example, a situation where the internal combustion engine 10 is new).

  The calculation of MFBax described above is performed under the condition that the same target air-fuel ratio is set for all the cylinders (◯, □, ☆, △), and under the operating condition in which the relationship shown in FIG. Done. Even if the same target air-fuel ratio is set, the air-fuel ratio realized in each cylinder (◯, □, ☆, △) occurs between cylinders, such as individual differences in the fuel injection valve 18 and individual differences in the intake system. It will be slightly different due to various differences. The slight difference in air-fuel ratio is projected on the MFBax based on the pre-stabilization output as a result of the air-fuel ratio feature of each cylinder. For this reason, the MFBax of each cylinder has a different value due to the air-fuel ratio feature of each cylinder.

  When the MFBax calculated for each cylinder (◯, □, ☆, Δ) is applied to the relationship shown in FIG. 12, the relationship shown in FIG. 13 can be specified. If the relationship shown in FIG. 13 is known, the air-fuel ratios realized in the respective cylinders with respect to the same target air-fuel ratio can be grasped from the air-fuel ratios indicated by ◯, □, ☆, and Δ. These air-fuel ratios represent the air-fuel ratio features of each cylinder. Therefore, the relationship shown in FIG. 13 can be grasped as a relative relationship of the air-fuel ratio feature of each cylinder.

  FIG. 14 is a diagram in which the relationship in FIG. 13 is replaced with the relationship between the injection correction amount and MFBax. As shown in FIG. 14, the relationship of the air-fuel ratio feature shown in FIG. 13 is that the injection correction amount to be given to each cylinder (◯, □, ☆, Δ) in order to make all the cylinders 12 have the same air-fuel ratio. Can be replaced by a realistic relationship.

  For example, FIG. 13 shows that the air-fuel ratio feature of the (◯) cylinder is richer than that of the (☆) cylinder. In other words, if the same fuel injection amount is commanded to both, the air-fuel ratio of the (◯) cylinder becomes richer than the air-fuel ratio of the (☆) cylinder. In this case, in order to make the air-fuel ratios coincide, it is necessary to make the injection correction amount given to the (◯) cylinder smaller than the injection correction amount given to the (☆) cylinder.

  The (□) cylinder has a richer air-fuel ratio feature than the (☆) cylinder. Therefore, in order to make the air-fuel ratios of both the same, it is necessary to make the injection correction amount given to the (□) cylinder smaller than the injection correction amount given to the (☆) cylinder. However, since the air-fuel ratio feature of the (□) cylinder is leaner than that of the (◯) cylinder, the injection correction amount given to the (□) cylinder must be larger than the injection correction amount given to the (◯) cylinder.

  These relative relationships appear correctly in the relationship shown in FIG. Thus, the relative relationship of the air-fuel ratio feature shown in FIG. 13 can be replaced with the relative relationship of the injection correction amount shown in FIG. If the relationship shown in FIG. 14 is prepared in advance, the injection correction amounts for all the other cylinders are determined based on the injection correction amount when the injection correction amount has been determined for at least one cylinder. Can do.

  FIG. 15 is a diagram in which the relationship shown in FIG. 14 is organized with reference to the injection correction amount of the (☆) cylinder. In this figure, the horizontal axis indicates the difference ΔMFBax between the MFBax of the (☆) cylinder and the MFBax of the other cylinders (○, □, △), and the vertical axis indicates the injection correction amount given to the (☆) cylinder and other A difference Δ injection correction amount from the injection correction amount to be given to the cylinder is shown.

  After the new internal combustion engine 10 (☆) starts operation, when the (☆) cylinder is first recognized as a stable cylinder, in the present embodiment, the (☆) cylinder is defined as a “representative cylinder”. In the representative cylinder (☆), the in-cylinder pressure sensor 20 has already output a stable output. For this reason, for the representative cylinder (☆), SA-CA10 is calculated based on the output of the in-cylinder pressure sensor 20 mounted therein, and the injection correction amount is determined so that the SA-CA10 approaches the target value. Can do.

  From the relationship shown in FIG. 15, the difference Δ injection correction amount between the injection correction amount given to the representative cylinder (☆) and the injection correction amount given to the other cylinders (◯, □, Δ) can be calculated. If the Δ injection correction amount obtained by the calculation is added to the injection correction amount of the representative cylinder (☆), the injection correction amount to be given to the other cylinders (◯, □, Δ) can be obtained.

  FIG. 16 is a diagram schematically showing the effect obtained by the above method. In the present embodiment, the ECU 40 starts air-fuel ratio feedback control using the above method. For this reason, in the present embodiment, air-fuel ratio feedback control based on the output of the in-cylinder pressure sensor 20 can be started for all the cylinders at the time t1st when the cylinder with the fastest convergence reaches the stable cylinder. As described above, according to the present embodiment, it is possible to shorten the time when the execution of the air-fuel ratio feedback control is prohibited due to the influence of the pre-stability change of the thermal strain characteristics as compared with the case of the first embodiment. .

[Operation of Embodiment 2]
17 and 18 are flowcharts of an example of a routine executed by the ECU 40 in the present embodiment. This routine is executed for each cycle of the internal combustion engine 10. When this routine is started, it is first determined whether or not the internal combustion engine 10 is new (step 120). The ECU 40 has, for example, a counter that counts the time from when the internal combustion engine 10 is first started, and determines that the internal combustion engine 10 is new until the count value reaches a determination value. The period in which the internal combustion engine 10 is determined to be new is limited to a very short period after the initial start. More specifically, the period is sufficiently shorter than the time required for the cylinder having the fastest convergence to converge to the stable cylinder, for example, several hours, one hour, several minutes, or the number of internal combustion engines 10 is several. It is set to a period of about 10 cycle operation.

  When it is determined that the internal combustion engine 10 is new, the processes of steps 122 to 126 are sequentially executed in order to grasp the air-fuel ratio feature of each cylinder. On the other hand, if it is determined that the internal combustion engine 10 is not new, it can be determined that all the cylinders 12 have already converged to stable cylinders. In this case, since it is no longer necessary to grasp the air-fuel ratio feature of each cylinder, the processing of steps 122 to 126 is jumped.

  As processing for grasping the air-fuel ratio feature of each cylinder, first, for each cylinder, the mass combustion ratio MFB based on the pre-stabilization output of the in-cylinder pressure sensor 20 is calculated over one cycle (step 122). The MFB of each cylinder is calculated by the same method as in step 100 shown in FIG.

  Next, the mass combustion ratio MFBax of the specific crank angle ax is read out for each cylinder (# 1 cylinder to # 4 cylinder) from the calculated value of MFB based on the pre-stabilization output (step 124).

  Next, the air-fuel ratio feature of each cylinder is grasped based on the MFBax of each cylinder based on the output before stabilization (step 126). In the present embodiment, the ECU 40 stores the relationship shown in FIG. 12, that is, the relationship between the MFBax and the air-fuel ratio established in the internal combustion engine 10. In step 126, specifically, the relative relationship of the air-fuel ratio realized in each cylinder is estimated by applying the MFBax of each cylinder to the relationship shown in FIG.

  The relationship shown in FIG. 12 is a relationship that is established under a specific operating state. Therefore, in order to grasp the air-fuel ratio feature of each cylinder based on this relationship, the MFBax applied thereto must also be obtained under the specific operating condition. In this routine, the MFBax is read from the MFB calculated in step 122. For this reason, in this embodiment, it is supposed that the process of the said step 122 will be performed under the specific driving | running state mentioned above.

  When the above process is completed, the process shown in FIG. 18 is executed. The process shown in FIG. 18 is a process for determining which cylinder is a stable cylinder and which cylinder is an unstable cylinder.

  Here, first, the mass combustion ratio MFBax of the specific crank angle ax is monitored for each cylinder (step 128). Specifically, the calculation of MFB over one cycle and the process of reading MFBax therefrom are executed for each cylinder.

  Next, a reference mass combustion ratio MFBax0 is calculated based on the engine speed NE and the engine load KL (step 130). The reference mass combustion rate MFBax0 is a mass combustion rate that should be originally generated at the specific crank angle ax, as in the first embodiment. Specifically, the processing in this step can be performed in the same manner as in step 104 shown in FIG.

Next, the ECU 40 determines whether the mass combustion ratio MFBax calculated for the # 1 cylinder is sufficiently close to the reference mass combustion ratio MFBax0. Specifically, it is determined whether or not the relationship of the following equation is established (step 132).
MFBax0−MFBax (1) ≦ judgment value (4 formulas)
Note that (1) included in the left side of the above (Expression 4) is the number of the cylinder that is processed in this step. Similarly, the numbers included in the displays of steps 136, 140, and 144 in FIG. 18 are the numbers of the cylinders that are processed by each step.

  The determination at step 132 is performed substantially in the same manner as the processing at step 106 shown in FIG. When this determination is positive, the ECU 40 determines that the # 1 cylinder has become a stable cylinder (step 134). On the other hand, if the above determination is negative, the processing of step 134 is jumped, and the subsequent processing proceeds while the # 1 cylinder is classified as an unstable cylinder.

  Thereafter, the same processing as in steps 132 and 134 is executed for each of the # 2, # 3, and # 4 cylinders (steps 136 to 146). With the above processing, each of the four cylinders 12 is classified as either a stable cylinder or an unstable cylinder. When these processes are completed, the process returns to FIG. 17 and the process for air-fuel ratio feedback control is started.

  Here, it is first determined whether or not there is a stable cylinder among the four cylinders (step 148). If it is determined that there is no stable cylinder, the current routine is terminated without starting the air-fuel ratio feedback control. As a result, the execution of the air-fuel ratio feedback control based on the pre-stable output of the in-cylinder pressure sensor 20 is prohibited, and the pre-stable output can be prevented from being inappropriately reflected in the operating state of the internal combustion engine 10.

On the other hand, when it is determined that a stable cylinder exists, an injection correction amount for the stable cylinder is calculated (step 150). Here, specifically, the following processes are sequentially executed.
1. The cylinder first determined as a stable cylinder is stored as a representative cylinder.
2. SA-CA10 is calculated for each stable cylinder. This calculation is performed based on the output of the in-cylinder pressure sensor 20 disposed in each stable cylinder.
3. The injection correction amount is calculated so that SA-CA10 approaches the target value for each stable cylinder.

Next, the injection correction amount for the unstable cylinder is calculated (step 152). Here, specifically, the following processes are sequentially executed.
1. Read out the air-fuel ratio features for each cylinder ascertained in step 126 (read out the relative relationship of the air-fuel ratios indicated by ◯, □, ☆, and Δ in FIG. 13).
2. The difference between the air-fuel ratio feature of the representative cylinder and the air-fuel ratio feature of each unstable cylinder is calculated (the difference between the air-fuel ratio indicated by ☆ and the air-fuel ratio indicated by ○, □, or Δ in FIG. 13 is calculated).
3. The difference between the air-fuel ratio feature of the representative cylinder and the air-fuel ratio feature of each unstable cylinder is converted into a difference in injection correction amount, that is, Δ injection correction amount (converted into Δ injection correction amounts indicated by ○, □, and Δ in FIG. 15) ).
4). Add the Δ injection correction amount to the injection correction amount for the representative cylinder to calculate the injection correction amount for each of the unstable cylinders.

  When the above processing ends, execution of air-fuel ratio feedback control based on the output of the in-cylinder pressure sensor 20 is permitted (step 154). According to the above processing, when one cylinder reaches a stable cylinder, air-fuel ratio feedback control of all cylinders is started with that cylinder as a representative cylinder. When the unstable cylinder shifts to the stable cylinder, the feedback control of the cylinder shifts from the control subordinate to the representative cylinder to the original feedback control. Therefore, according to the present embodiment, air-fuel ratio feedback control with higher accuracy than in the case of the first embodiment during the period from when the internal combustion engine 10 is initially started until all cylinders reach stable cylinders. Can be provided.

[Modification of Embodiment 2]
Incidentally, the second embodiment is the same as the first embodiment in that the following modifications can be applied.
1. It is possible to determine whether or not each cylinder has shifted to a stable cylinder based on the difference between the previous value MFBax1 and the current value MFBax.
2. As a process when all cylinders are unstable cylinders, feedback gain reduction can be used instead of control inhibition.
3. Whether or not each cylinder has shifted to a stable cylinder can be determined from the output itself of the in-cylinder pressure sensor 20 instead of the mass combustion ratio MFBax.
4). The specific crank angle ax is not limited to ATDC 90 ° CA, and may be in the region from the crank angle at which the maximum in-cylinder pressure is generated to the crank angle at which the exhaust valve 32 is opened. A point that is closer to the corner.

  Further, in the second embodiment described above, the process for grasping the air-fuel ratio feature of the cylinder 12 is executed only when the internal combustion engine 10 is new, but the present invention is not limited to this. For example, in the internal combustion engine 10, all the cylinders that have shifted to the stable cylinder may return to the unstable cylinder due to the deposit cleaning process. In consideration of changes over time of the individual cylinders 12, it is desirable to grasp the air-fuel ratio feature for each cylinder again under such circumstances. For this reason, the process for grasping the air-fuel ratio feature may be executed immediately after all the cylinders return from the stable cylinders to the unstable cylinders.

  In the second embodiment described above, the air-fuel ratio of each cylinder specified from the relationship shown in FIG. 13 is grasped as the air-fuel ratio feature (see step 126), and this is converted into a Δ injection correction amount (see step 152). doing. However, the air-fuel ratio feature of each cylinder is not limited to this. That is, the injection correction amount for each cylinder specified from the relationship shown in FIG. 14 may be grasped as an air-fuel ratio feature and converted to a Δ injection correction amount.

  In the second embodiment described above, whether or not each cylinder has reached a stable cylinder is determined based on the output (mass combustion ratio MFBax) of the in-cylinder pressure sensor 20. However, the determination method is not limited to this, and can be replaced with the method of the third embodiment (method based on the deposit accumulation amount) described below.

Embodiment 3 FIG.
Next, Embodiment 3 of the present invention will be described with reference to FIGS. The system of the present embodiment can be realized using the hardware shown in FIG.

[Features of Embodiment 3]
In the first and second embodiments described above, it is determined whether or not the pre-stabilization change of the thermal strain characteristic of the pressure-sensitive portion 22 has settled based on whether or not the mass combustion ratio MFBax of the specific crank angle ax has reached a stable value. We are going to judge. By the way, the pre-stabilization change of the thermal strain characteristic is converged by depositing sufficient deposits on the pressure sensitive part 22 as described above. For this reason, it is also possible to determine whether or not the change has been settled based on the amount of deposit attached to the pressure-sensitive portion 22. The present embodiment is characterized in that the determination is made based on the deposit adhesion amount instead of the mass combustion ratio MFBax.

[Deposit amount and output drift]
The upper part of FIG. 19 shows a state in which the load of the internal combustion engine 10 suddenly increases at time t1 with the return from the fuel cut. On the other hand, the lower part of FIG. 19 shows a state in which the output of the in-cylinder pressure sensor 20 drifts with a sudden increase in load. FIG. 20 shows the result of extracting only the change due to the output drift from the waveform in the lower part of FIG.

  When the fuel cut is restored, a large temperature change occurs in the cylinder 12. Similarly, a large temperature change occurs in the cylinder 12 when the fuel cut is performed. When such a temperature change occurs, a drift as shown in FIG. 20 occurs in the output of the in-cylinder pressure sensor 20. The ECU 40 can detect the maximum drift amount ΔP1 by monitoring the output of the in-cylinder pressure sensor 20 before and after the return from the fuel cut or before and after the execution of the fuel cut.

  The maximum drift amount ΔP1 increases as the temperature change in the cylinder 12 is easily transmitted to the pressure-sensitive portion 22. As described above, the deposit attached to the pressure-sensitive portion 22 has a function of blocking heat. For this reason, the maximum drift amount ΔP1 appearing in the output of the in-cylinder pressure sensor 20 decreases as the deposit adhesion amount increases.

  FIG. 21 shows the relationship between the operating state of the internal combustion engine 10 and the output drift amount of the in-cylinder pressure sensor 20. A straight line indicated by ΔPα in the figure represents the maximum drift amount (hereinafter referred to as “reference drift amount”) that occurs under the condition that no deposit is attached. On the other hand, ΔP1 is the actual maximum drift amount that is smaller than ΔPα due to the deposit.

  As shown in FIG. 21, the maximum drift amounts ΔPα and ΔP1 have different values depending on the operating state of the internal combustion engine 10 when the load suddenly changes. Specifically, the operation state after returning from the fuel cut or the operation state before starting the fuel cut becomes larger as the load becomes higher.

In the present embodiment, the ECU 40 stores a map of ΔPα shown in FIG. 21, that is, a map that defines the reference drift amount ΔPα in relation to the operating state of the internal combustion engine 10. According to this map, when a sudden change in load occurs, the reference drift amount ΔPα corresponding to the operation state at that time can be specified. Then, when the maximum drift amount ΔP1 actually generated with sudden load change is used, the drift reduction rate indicated by the following equation can be obtained.
Drift reduction rate = (ΔPα−ΔP1) / ΔPα (Expression 5)

  This drift reduction rate correctly represents how much the actual maximum drift amount ΔP1 is decreasing with respect to the reference drift amount regardless of the operating state of the internal combustion engine 10. Since the maximum drift amount ΔP1 indicates attenuation according to the deposit adhesion amount, a correlation as shown in FIG. 22 is recognized between the drift reduction rate and the deposit adhesion amount. For this reason, the drift reduction rate indicated by the above (Formula 5) can be treated as a characteristic value of the deposit adhesion amount.

[Operation of Embodiment 3]
As described above, the system according to the present embodiment determines whether or not the pre-stabilization change of the thermal strain characteristic of the pressure-sensitive portion 22 has been settled based on the deposit amount. More specifically, this system makes the above determination based on a drift reduction rate that can be treated as a characteristic value of the deposit adhesion amount.

  FIG. 23 is a flowchart of an example of a routine executed by the ECU 40 in the present embodiment. The routine shown in FIG. 23 is started when the load of the internal combustion engine 10 changes suddenly. More specifically, this routine is started when the fuel cut is started and when the fuel cut is restored.

  When this routine is started, first, the maximum drift amount ΔP1 is calculated based on the output of the in-cylinder pressure sensor 20 (step 156).

  Next, a reference drift amount ΔPα corresponding to the current operating state is calculated according to a map of ΔPα as shown in FIG. 21 (step 158).

Next, determination processing based on the drift reduction rate is executed (step 160). Specifically, the following two processes are executed.
1. The drift reduction rate is calculated by applying the maximum drift amount ΔP1 and the reference drift amount ΔPα to the above equation (5).
2. Determine if the drift reduction rate is greater than or equal to the criterion value.

  The drift reduction rate increases as the deposit amount on the in-cylinder pressure sensor 20 increases. On the other hand, the pre-stabilization change of the thermal strain characteristic of the pressure-sensitive portion 22 also converges as the deposit accumulation amount increases. The determination value in step 160 is a value set based on a drift reduction rate that occurs when a deposit sufficient to converge the pre-stability change is accumulated on the in-cylinder pressure sensor 20.

  For this reason, if the determination in step 160 is negative, it can be determined that the change in stability of the thermal strain characteristics has not yet settled. In this case, the current routine is immediately terminated. As a result, the state where the control based on the output of the in-cylinder pressure sensor 20 is prohibited is maintained.

  On the other hand, if the determination in step 160 is affirmative, it can be determined that the pre-stabilization change of the thermal strain characteristics has subsided. In this case, the ECU 40 permits the execution of the control based on the output of the in-cylinder pressure sensor 20 through the processing of steps 108 and 110.

  According to the above processing, as in the case of the first embodiment, before the output of the in-cylinder pressure sensor 20 is stabilized, the output is prevented from being inappropriately reflected in the operating state of the internal combustion engine 10. be able to. As described above, the system of the present embodiment can achieve the same effect as the system of the first embodiment.

[Modification of Embodiment 3]
Incidentally, the third embodiment is similar to the first embodiment in that feedback gain reduction can be used in place of prohibiting control as a process when all cylinders are unstable cylinders.

  In the above-described third embodiment, the deposit accumulation state is determined based on the maximum drift amount ΔP1 appearing in the output of the in-cylinder pressure sensor 20. However, the method for determining the deposit state is not limited to this. This method can be replaced with a known method as disclosed in, for example, Japanese Patent Application Laid-Open No. 2014-95368.

DESCRIPTION OF SYMBOLS 10 Internal combustion engine 12 Cylinder 16 Spark plug 18 Fuel injection valve 20 In-cylinder pressure sensor 22 Pressure sensing part 32 Exhaust valve 40 Control unit
ax Specific crank angle
MFB mass combustion rate
MFBax Mass combustion ratio of specific crank angle
Mass combustion rate based on MFBax0
Pc In-cylinder pressure
Pc0 True value of in-cylinder pressure
Pcax specific crank angle output
MFBax1 Previous value of mass combustion ratio at specified crank angle ΔP1 Maximum drift amount ΔPα Standard drift amount

Claims (12)

An in-cylinder pressure sensor having a pressure-sensitive portion located in a cylinder of the internal combustion engine;
A control unit for controlling the actuator of the internal combustion engine based on the output of the in-cylinder pressure sensor,
The control unit is
Based on the output of the in-cylinder pressure sensor, it is determined whether or not the pre-stabilization change of the thermal strain characteristics of the pressure sensitive part has been settled,
Until the determination is affirmed, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is suppressed to be lower than the degree of reflection used after the determination is affirmed. . The control unit is
Based on the output of the in-cylinder pressure sensor, the mass combustion ratio MFBax at a specific crank angle ax belonging to the region from the maximum in-cylinder pressure generation crank angle to the exhaust valve opening crank angle is calculated for the cylinder,
Determine whether the mass combustion rate MFBax has reached a stable value,
2. The control device according to claim 1, wherein when it is determined that the mass combustion ratio MFBax has reached a stable value, it is determined that the pre-stability change of the thermal strain characteristics has been settled. The control unit is
For the cylinder, it is determined whether or not the output Pcax of the in-cylinder pressure sensor at a specific crank angle ax belonging to the region from the maximum in-cylinder pressure generation crank angle to the exhaust valve opening crank angle has reached a stable value,
2. The control device according to claim 1, wherein when it is determined that the output Pcax has reached a stable value, it is determined that the pre-stabilization change of the thermal strain characteristic has subsided.   The control device according to claim 2 or 3, wherein the specific crank angle ax belongs to a rear 1/4 region in a region from a maximum in-cylinder pressure generating crank angle to an exhaust valve opening crank angle. The control unit is
When the difference between the reference mass combustion rate MFBax0 to be generated at the specific crank angle ax and the mass combustion rate MFBax is equal to or less than a determination value, it is determined that the mass combustion rate MFBax has reached a stable value. The control device according to any one of claims 2 to 4. The control unit is
When the difference between the past mass combustion ratio MFBax1 obtained in the cylinder and the mass combustion ratio MFBax is equal to or less than a determination value, it is determined that the mass combustion ratio MFBax has reached a stable value,
The past mass combustion rate MFBax1 was calculated under conditions where it can be considered that the same mass combustion rate occurs at the specific crank angle ax as compared to the condition of the cylinder when the mass combustion rate MFBax was calculated. The control device according to any one of claims 2 to 4, wherein the control device is a latest latest value. The control unit is
Calculate the amount of drift that appears in the output of the in-cylinder pressure sensor with the on / off switching of the fuel cut,
When the difference between the reference drift amount and the drift amount to be generated when no deposit is attached to the pressure-sensitive part under the condition in which the drift amount has occurred and the thermal strain characteristic is stable The control device according to claim 1, wherein it is determined that the previous change has subsided. The control of the actuator is
State determination based on the output of the in-cylinder pressure sensor;
A process for switching a command value calculation method for the actuator according to a result of the state determination;
Supplying the command value to the actuator, and
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has been affirmed, by prohibiting the execution of the state determination, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is determined. The control device according to claim 1, wherein the control device is suppressed to be lower than a degree of reflection used after affirmation. The control of the actuator includes feedback control that reflects a correction amount based on the output of the in-cylinder pressure sensor in a command value for the actuator,
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has been affirmed, the feedback gain of the correction amount is made smaller than the gain used after the determination is affirmed, so that the in-cylinder pressure sensor 9. The control device according to claim 1, wherein a degree to which an output is reflected in the control is suppressed to be lower than a degree of reflection used after affirmation of the determination. The control of the actuator includes feedback control that reflects a correction amount based on the output of the in-cylinder pressure sensor in a command value for the actuator,
The control unit is
Until the determination that the pre-stability change of the thermal strain characteristic has settled is affirmed, by prohibiting the execution of the feedback control, the degree to which the output of the in-cylinder pressure sensor is reflected in the control is determined. The control device according to claim 1, wherein the control device is suppressed to be lower than a degree of reflection used after affirmation. The internal combustion engine is an internal combustion engine having a plurality of cylinders;
The in-cylinder pressure sensor is arranged for each cylinder,
The actuator includes a fuel injection valve arranged for each cylinder,
The feedback control includes air-fuel ratio feedback control that causes the fuel injection valve to inject a fuel injection amount to which an output of the in-cylinder pressure sensor is fed back,
The control unit is
For each cylinder, based on the output of the in-cylinder pressure sensor before it is determined that the pre-stability change of the thermal strain characteristic has been settled, the air-fuel ratio feature of the cylinder is grasped,
When it is determined that the pre-stability change of the thermal strain characteristic has been settled in at least one cylinder,
Based on the output of the in-cylinder pressure sensor of the stable cylinder for which the determination has been made, the injection correction amount of the stable cylinder is calculated,
From the relative relationship between the air-fuel ratio feature of an unstable cylinder that has not been determined that the pre-stability change of the thermal strain characteristic has subsided and the air-fuel ratio feature of the stable cylinder, the relative relationship between the injection correction amounts of both cylinders is specified. ,
By calculating the injection correction amount of the unstable cylinder by applying the relative relationship of the injection correction amount to the injection correction amount of the stable cylinder,
The air-fuel ratio feedback control is performed by reflecting the injection correction amount of the stable cylinder in the fuel injection amount of the stable cylinder and reflecting the injection correction amount of the unstable cylinder in the fuel injection amount of the unstable cylinder. The control device according to claim 9, wherein the control device is executed. The control unit is
If the determination that the pre-stabilization change of the thermal strain characteristic has been settled is denied for all the cylinders, then immediately grasp the air-fuel ratio features of all the cylinders,
12. The method according to claim 11, wherein when a plurality of cylinders are determined to be the stable cylinders, an injection correction amount of the cylinder determined to be the stable cylinder earliest is used as a basis for the injection correction amount of the unstable cylinder. The control device described.

Images