JP6107150B2 - Control device for internal combustion engine - Google Patents

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 can correct an ignition timing.

  Patent Document 1 discloses an internal combustion engine including an in-cylinder pressure sensor and a crank angle sensor. From the signals output from these sensors, the combustion ratio of each cylinder in the combustion cycle can be grasped. Then, the crank angle at which the combustion ratio becomes 50% (hereinafter referred to as CA50) is grasped. Furthermore, in the technique described in Patent Document 1, first, an optimum CA 50 is set for outputting the target torque. The ignition timing corresponding to the optimum CA50 is set as a target value, and control is performed to bring the actual ignition timing closer to this target value. By performing this control, a stable combustion cycle can be performed. As a result, a target torque can be obtained.

  Specific contents of the control include ignition timing feedback control (hereinafter referred to as FB control). In this FB control, a correction value is calculated from the difference between the optimum CA50 and the actually measured CA50. Based on this correction value, the next ignition timing is advanced or retarded, whereby the actual ignition timing can be brought close to the target value. By this FB control, it is possible to suppress the actual ignition timing from constantly deviating from the target value. Moreover, long-term combustion fluctuations can also be suppressed.

JP 2009-079594 A Japanese Patent Laid-Open No. 5-321773 JP 2006-312919 A

  By the way, when idling in an internal combustion engine or in a situation where combustion is likely to be relatively unstable (lean combustion, a state with a high EGR rate, etc.), a combustion fluctuation with a short period that coincides with the combustion cycle is likely to occur. In these situations, combustion fluctuations with a short cycle cannot be suppressed only by performing the above-mentioned FB control. As a result, this short cycle combustion fluctuation may cause drivability deterioration.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a control device for an internal combustion engine that can suppress a short-cycle combustion fluctuation.

In order to achieve the above object, a first invention is a control device for an internal combustion engine,
An in-cylinder pressure sensor;
In a control device for an internal combustion engine comprising a crank angle sensor,
Combustion state parameter calculating means for calculating a combustion state parameter whose value is determined for each combustion cycle in accordance with the relationship between the in-cylinder pressure and the crank angle;
Target value setting means for setting a target combustion state parameter that is a target value of the combustion state parameter;
When the combustion state parameter shows a value obtained when combustion is early with respect to the target combustion state parameter, the ignition timing is advanced, and when the combustion state parameter is late with respect to the target combustion state parameter and ignition timing correcting means for retarding the ignition timing when the shows the values obtained, Ru comprising a.
Here, the ignition timing correction means determines the correction amount of the ignition timing in proportion to the difference between the target combustion state parameter and the combustion state parameter .
In the present invention, the control device for the internal combustion engine further includes:
A measuring means for measuring the number of times of combustion in the combustion cycle of each cylinder;
Combustion number determination means for determining whether or not the number of combustion times is a certain number of times;
Combustion state parameter change amount calculating means for calculating a combustion state parameter change amount that is a change amount with respect to the combustion state parameter stored immediately before the combustion state parameter;
Data storage means for storing the combustion state parameter stored immediately before the combustion state parameter and the combustion state parameter change amount as data in association with each other;
A correlation coefficient calculating means for calculating a correlation coefficient between the combustion state parameter stored in the data storage means and the combustion state parameter change amount when the number of data in the data storage means exceeds a certain number;
When the number of combustions is equal to or greater than a certain number and the correlation coefficient is equal to or greater than a certain value, the data storage means has a plane set with the combustion state parameter and the combustion state parameter variation as an axis. A distribution area calculation means for calculating a distribution area of stored data;
When the distribution area calculated by the distribution area calculation means is smaller than the distribution area calculated immediately before, the proportional gain of the correction amount of the ignition timing with respect to the difference between the target combustion state parameter and the combustion state parameter is increased, Proportional gain calculation means for reducing the proportional gain in the above case.

The second invention is the first invention, wherein
The combustion state parameter is a specific combustion rate point that is a crank angle at a specific combustion rate of each cylinder,
The ignition timing correction means corrects the ignition timing to the retard side when the difference between the specific combustion ratio point and a target specific combustion ratio point that is a target value of the specific combustion ratio point is a positive number, and the specific combustion When the difference between the ratio point and the target specific combustion ratio point is a negative number, the ignition timing is corrected to the advance side.

The third invention is the first or second invention, wherein
Average processing means for performing an average process on the combustion state parameter to obtain an average post-combustion state parameter;
Feedback control means for performing feedback control of ignition timing using a difference between the target combustion state parameter and the average post-combustion state parameter;
It is characterized by providing.

In a fourth aspect based on the third aspect , the feedback control means uses any one of P control, PI control, and PID control for a difference between the target combustion state parameter and the average post-average combustion state parameter. The ignition timing is corrected based on the correction value obtained in this manner.

In addition, a fifth aspect of the present invention is any one of the first to fourth aspects of the invention,
Combustion state parameter storage means for storing the combustion state parameter;
Combustion state parameter change amount calculating means for calculating a combustion state parameter change amount that is a change amount with respect to the combustion state parameter stored immediately before the combustion state parameter;
Data storage means for storing the combustion state parameter stored immediately before the combustion state parameter and the combustion state parameter change amount as data in association with each other;
A correlation coefficient calculating means for calculating a correlation coefficient between the combustion state parameter stored in the data storage means and the combustion state parameter change amount when the number of data in the data storage means exceeds a certain number;
Prohibiting means for prohibiting correction of the ignition timing by the ignition timing correcting means when the correlation coefficient is smaller than a certain value;
It is characterized by providing.

  According to the first invention, the phase of the next ignition timing is controlled depending on whether the combustion is early or late. Thereby, the fluctuation | variation of the in-cylinder pressure change can be suppressed and the combustion fluctuation | variation which generate | occur | produces for every combustion cycle can be suppressed.

  According to the second aspect of the invention, the advance or retard angle of the next ignition timing can be determined by calculating using the specific combustion ratio point as a parameter.

According to the first invention, the correction amount of the ignition timing is determined based on the difference (error) from the target value. For this reason, the optimal correction amount at that time can be calculated for the combustion fluctuation that is currently occurring.

According to the third and fourth inventions, the combustion state parameter can be brought close to the target value by correcting the next ignition timing in parallel with the feedback control.

According to the fifth aspect , by using the correlation coefficient, it can be determined whether or not a combustion fluctuation that occurs every combustion cycle occurs. For this reason, when correcting the ignition timing, a more effective state can be grasped. As a result, the ignition timing correction accuracy is improved.

According to the first invention, when the distribution area is smaller than the distribution area calculated immediately before, the proportional gain is changed to a larger value, and the influence of the ignition timing error can be more reflected in the correction value. Become. As a result, the ignition timing correction accuracy is improved. On the other hand, when the distribution area is equal to or greater than the distribution area calculated immediately before, the proportional gain can be changed to a smaller value to suppress data variation. As a result, the accuracy of prediction from the current combustion to the next combustion can be improved.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic block diagram for demonstrating the structure of the system of Embodiment 1 of this invention. The block diagram of the control system structure in this embodiment is shown. An example of the control result of FB control is shown. The result of having investigated about the fluctuation | variation of the measured value at the time of fixed ignition is represented. It is a figure showing consideration based on the investigation result in FIG. It shows the fluctuation of the measured value when the present invention is implemented. It is the figure which showed the effect which this invention has on illustrated torque fluctuation. It is a figure showing the fluctuation | variation of CA50 in each cylinder. It is a figure showing the tendency of CA50 at the time of fixed ignition. It is a figure showing the change of the distribution range of data of CA50 and the variation | change_quantity of CA50. 4 is a flowchart of an ignition timing control routine executed by an ECU 30 in the present embodiment. In the embodiment of the present invention, it is a flow chart of a routine for determining whether or not to execute a next combustion prediction control executed by the ECU 30. 4 is a flowchart of a proportional gain determination routine executed by an ECU 30 in the embodiment of the present invention.

Embodiment 1 FIG.
[System Configuration of Embodiment 1]
FIG. 1 is a schematic configuration diagram for explaining the configuration of the system according to the first embodiment of the present invention. The system shown in FIG. 1 includes an engine 10. Normally, the engine 10 is composed of a plurality of cylinders, but only one cylinder is depicted in FIG. In the present invention, the number of cylinders and the cylinder arrangement are not limited thereto.

  The engine 10 is provided with a piston 12. The piston 12 is connected to a crankshaft (not shown). A crank angle sensor 16 is provided on the crankshaft. An in-cylinder pressure sensor 20 is attached to the combustion chamber 18 of the engine 10 to detect the in-cylinder pressure. Further, an ignition plug 22 and a fuel injection valve 24 are attached to the combustion chamber 18. Further, the combustion chamber 18 is provided with an intake valve 26 and an exhaust valve 28. In the present invention, the position of the fuel injection valve 24 is not limited to this. For example, a fuel injection valve may be attached to the intake port. Alternatively, it may be attached to both the combustion chamber 18 and the intake port.

The system configuration of the first embodiment includes an ECU (Engine Control Unit) 30 that controls the operating state of the engine 10. Various sensors such as the crank angle sensor 16 and the in-cylinder pressure sensor 20 are connected to the input side of the ECU 30. These various sensors detect information for controlling the engine 10 and output the detected information to the ECU 30 as signals. Specifically, the crank angle sensor 16 outputs a pulse signal synchronized with the rotation of the crankshaft. The in-cylinder pressure sensor 20 outputs a signal corresponding to the in-cylinder pressure that changes depending on the combustion cycle.

  ECU30 detects the driving | running state of the engine 10 based on the signal which said various sensors output. Specifically, the crank angle (CA) is detected from a constant pulse signal output from the crank angle sensor 16. In-cylinder pressure (P) is detected from the signal output from in-cylinder pressure sensor 20. Then, the ECU 30 calculates the combustion ratio in each cylinder from the crank angle and the in-cylinder pressure. On the other hand, actuators such as a spark plug 22 and a fuel injection valve 24 are connected to the output side of the ECU 30. The ECU 30 sets the ignition timing by supplying a drive signal to the spark plug 22.

  FIG. 2 shows a block diagram of the control system configuration in the present embodiment. FIG. 2 shows the engine 10 and the ECU 30 shown in a dashed-dotted frame. The ECU 30 is provided with a target ignition timing calculation unit 32, a target CA50 calculation unit 34, a CA50 calculation unit 48, an annealing processing unit 36, a PI control unit 40, and a next ignition timing calculation unit 50.

In the present embodiment, FB control of the ignition timing is performed in order to output an optimum torque in the engine 10. The FB control, with respect to the target ignition timing calculated by the target ignition timing calculation unit 32, the correction value [Delta] SA _PI is added. With reference to FIG. 2, illustrating the flow of the correction value [Delta] SA _PI in the FB control is calculated. First, the CA50 calculation unit 48 receives signals from a crank angle sensor 16 (not shown) provided in the engine 10 and an in-cylinder pressure sensor 20 (not shown) provided in the engine 10. Based on these signals, the actually measured CA50 (hereinafter referred to as an actual measurement value) is calculated. The actual measurement value output from the CA50 calculation unit 48 is subjected to a smoothing process (hereinafter also referred to as an average process) by the smoothing processing unit 36. Then, a difference between the actually measured value subjected to the annealing process and the target value output from the target CA50 calculation unit 34 is calculated. This difference is PI control in the PI control unit 40, the correction value [Delta] SA _PI. Then, at the next ignition timing calculation unit 50, it is added to the correction value [Delta] SA _PI target ignition timing, the next ignition timing is calculated. The unit of the target ignition timing is BTDC, the unit of the correction value [Delta] SA _PI is ATDC.

[Ignition timing control in this embodiment]
FIG. 3 shows an example of the control result of the FB control. In FIG. 3, the vertical axis indicates the phase of CA50, and the horizontal axis indicates the change in time. Further, in FIG. 3, the target value output by the target CA50 calculation unit 34 is indicated by a one-dot chain line. Further, the left diagram in FIG. 3 shows the waveform of the CA 50 before the FB control is performed. In the left diagram of FIG. 3, the waveform that fluctuates in a short cycle is a waveform that indicates an actual measurement value. Thus, the actual measurement value fluctuates in a short time period due to the fluctuation of the combustion speed for each combustion cycle. If the FB control is performed on the actual measurement value as it is, the control is diverged. For this reason, the average processing is performed on the actually measured values by the annealing processing unit 36. The left diagram in FIG. 3 shows the waveform after the averaging process. In the waveform after the averaging process, fluctuations in a short period are excluded. And the right figure of FIG. 3 shows the waveform of CA50 after the FB control is performed. The right diagram in FIG. 3 shows a state in which the waveform of CA50 is close to the target value as a result of the FB control. As described above, when the FB control of the ignition timing is performed, the constant deviation of the CA50 from the target value is corrected. In the present specification, “short-term fluctuation” refers to fluctuations in measured values that occur in each combustion cycle.

  However, as shown by the waveform of CA50 in the right diagram of FIG. 3, the waveform of CA50 after execution of the FB control has a short-cycle waveform that occurs every combustion cycle. This is because the FB control is performed in a state where the averaging process is performed to suppress a short period fluctuation. There is a concern that this short period fluctuation may cause a deterioration in drivability.

  Therefore, the ignition timing control in the present embodiment uses a method of performing new control in addition to the FB control in order to suppress the fluctuation of the short cycle.

  FIG. 4 shows the result of examining the fluctuation of the actual measurement value at the time of fixed ignition. The fixed ignition means that the ignition timing is not corrected. As shown by the waveform of CA50 in FIG. 4, at the time of fixed ignition, it was confirmed that the advance angle and the retard angle of the actually measured value were alternately repeated for each combustion cycle.

  FIG. 5 is a diagram showing consideration based on the above-described investigation results. FIG. 5 shows a state in which the actual measurement value this time is shifted to the retard side from the target value. Here, based on the above investigation results, it can be predicted that the next actual measurement value in FIG. However, if the FB control is performed in this state, the next ignition timing is over-advanced as indicated by the FB control point in FIG. In this case, the variation of CA50 is increased compared to the fixed ignition. As a result, the FB control diverges. Therefore, in the present invention, the next combustion prediction control is performed that predicts whether the next actual measurement value changes from the actual measurement value to the advance angle or the retard angle, and controls the ignition timing in a phase opposite to the direction of the change. I decided to do it. When this control is performed, the actually measured value can be optimally corrected with respect to the target value as shown by the point of the present invention in FIG. In this way, in the present invention, first, it is grasped whether the actual measurement value of this time is shifted to the advance side or the retard side with respect to the target value. Further, it is predicted whether the next actual measurement value is shifted to the advance side or the retard side with respect to the target value. Then, in order to correct the deviation from the predicted target value, the next ignition timing is corrected to the advance side or the retard side. Moreover, the new control method in this embodiment is adding correction in the direction opposite to FB control in addition to FB control. Thereby, it is possible to suppress the excessive advance angle and the excessive retard angle, and to suppress a short period fluctuation.

FIG. 2 shows a configuration for realizing new control in the present embodiment. In FIG. 2, a system configuration relating to new control is shown in a dotted frame indicated by A. In this frame A, the difference between the actually measured value and the target value output from the target CA50 calculation unit 34 is calculated. Then, this difference is P controlled by the P control unit 44 provided in the ECU 30, the correction value [Delta] SA _P is calculated. The unit of the correction value [Delta] SA _P is ATDC.

Further, in the FIG. 2 B, and the correction value [Delta] SA _PI calculated by PI control unit 40, the difference in a correction value [Delta] SA of the correction value [Delta] SA _P calculated by the P control unit 44 is calculated as a control deviation. Next, in the next ignition timing calculation unit 50, the correction value ΔSA is added to the target ignition timing, and the next ignition timing is calculated.

Here, it will be described in detail correction value ΔSA _P. Correction value [Delta] SA _P is calculated based on the difference between the measured value and the target value. For example, when the difference between the actual value and the target value is a positive number, that is, the actual value is on the retard side with respect to the target value. In this case, the correction value ΔSA _P is a positive number. The correction value [Delta] SA _P is (adds BTDC) for retarding the next ignition timing acting in the direction. On the other hand, when the difference between the actual measurement value and the target value is a negative number, that is, the actual measurement value is on the advance side with respect to the target value. In this case, the correction value ΔSA _P becomes a negative number. The correction value [Delta] SA _P is (subtracts the BTDC) of the next ignition timing advance angularly acting in the direction.

The correction value [Delta] SA _P and P control is performed on the difference between the measurement value and the target value in the process of being calculated, the proportional gain is used. The higher the proportional gain, the greater the correction value [Delta] SA _P. On the other hand, as the proportional gain is small, the correction value [Delta] SA _P is also reduced.

  FIG. 6 shows the fluctuation of the actual measurement value when the present invention is implemented. Compared to the variation shown in FIG. 4, it can be seen that in FIG. 6, the variation in CA50 in which the advance angle and the retard angle are alternately repeated is suppressed. As a result of suppressing this short period fluctuation, drivability is improved.

  FIG. 7 is a diagram showing the effect of the present invention on the indicated torque fluctuation. CA50 is correlated with the indicated torque. For this reason, when CA50 fluctuates in a short cycle, the indicated torque also fluctuates accordingly. As indicated by a circle in FIG. 7, the fluctuation in the vicinity of the explosion frequency is reduced in the present invention as compared with the FB control. Thus, by reducing the fluctuation in the indicated torque, the idling speed can be reduced, and the fuel consumption can be improved. In addition, since the degree of freedom for setting the idling speed is increased, the controllability of the S & S stop control can be improved. S & S here means stop & start. Specifically, the control is such that idling of the engine 10 is temporarily stopped and restarted.

  FIG. 8 is a diagram showing the variation of CA50 in each cylinder. First, the left figure of FIG. 8 represents the change of CA50 in each cylinder before FB control. This figure shows a state in which the waveforms of the cylinders (# 1- # 4) are far away from the target values. Next, the center diagram of FIG. 8 shows the variation of CA50 in each cylinder after the FB control is performed. This figure shows a state in which the constant deviation and long-term fluctuation of the CA50 waveform in each cylinder are suppressed. And the right figure of FIG. 8 represents the fluctuation | variation of CA50 in each cylinder after implementation of this invention. This figure shows a state in which fluctuations in the short cycle of CA50 in each cylinder are suppressed. Thus, torque fluctuations can be suppressed by the CA50 of each cylinder converging to the target value.

[Next combustion prediction control execution feasibility judgment method]
The effect of the next combustion prediction control can be obtained by performing the combustion fluctuation in a short cycle, that is, when the combustion fluctuation in which the advance angle and the retard angle are alternately repeated is generated. On the other hand, when such combustion fluctuation does not occur, the effect is small. For this reason, it is preferable to perform the next combustion prediction control after determining whether such a fluctuation has occurred. Here, a method for determining whether or not to execute the next combustion prediction control will be described with reference to FIG.

  FIG. 9 is a diagram showing the tendency of CA50 at the time of fixed ignition. In FIG. 9, the horizontal axis indicates the previous measured value of CA50, and the vertical axis indicates the amount of change in CA50. The amount of change in CA50 is the difference between the current measured value and the previous measured value. The data plotted in FIG. 9 is data in which the current measured value and the amount of change in CA 50 at the time when the current measured value is measured are associated with each other and stored. A correlation coefficient is calculated based on this data. If the correlation coefficient shows a numerical value greater than or equal to a certain value, it can be understood that a variation in which the advance and retard of the ignition timing are alternately repeated is caused in the combustion cycle. By performing this determination method, the correction accuracy of the next combustion prediction control can be improved.

The calculation of the correlation coefficient may be realized by the following equation (1). In the following formula 1, r is a correlation coefficient, and n is the number of data. Further, x is CA50, and y is the amount of change in CA50.

[Proportional gain determination method]
In the next combustion prediction control, when calculating the correction value [Delta] SA _P obtained from the P control, the proportional gain is used. Here, a method for setting the proportional gain to an optimum value using data collected in the method for determining whether or not to execute the next combustion prediction control will be described with reference to FIG. FIG. 10 shows elements common to the items of the respective axes shown in FIG.

FIG. 10 is a diagram illustrating changes in the distribution range of data of CA50 and the amount of change in CA50. In FIG. 10, data distribution ranges A, B, and C are displayed by respective frame lines. These distribution ranges are expressed as distribution areas on a plane having the CA50 variation on the vertical axis and the CA50 on the horizontal axis. According to the data plot shown in FIG. 10, it is determined that the distribution range is A. The distribution range of A is the smallest distribution range compared to B and C. For this reason, when the collected data is in the distribution range of A, the proportional gain is set to a larger value. As a result, the deviation of the target value from the measured value is reflected in the more correction value [Delta] SA _P, it is possible to improve the correction accuracy. Further, for example, when the collected data is C indicating the maximum distribution range, the proportional gain is set to a smaller value. As a result, a fluctuation in which the advance angle and the retard angle are alternately repeated is likely to occur, and the data distribution range is reduced. For this reason, the prediction accuracy of the next actual measurement value can be improved. Thus, the proportional gain is set to a smaller value as the distribution area increases, and the proportional gain is set to a larger value as the distribution area decreases.

  As described above, there are the following reasons for comparing the distribution areas. This is because even if there is a correlation in the data, if the distribution area is large, it cannot be said that short-period fluctuations occur, and it is not preferable to add a proportional gain. For this reason, it is preferable to perform a two-stage determination of the correlation coefficient and the distribution area.

  The ignition timing control in the present embodiment described above is realized by the ECU 30 executing the following ignition timing control routine.

[Ignition timing control routine]
FIG. 11 is a flowchart of an ignition timing control routine executed by the ECU 30 in the present embodiment. In the routine shown in FIG. 11, first, in-cylinder pressure and crank angle are read (S100). Specifically, the ECU 30 detects the in-cylinder pressure and the crank angle based on signals output from the in-cylinder pressure sensor 20 and the crank angle sensor 16.

  Next, CA50 is calculated (S102). Specifically, the CA50 calculation unit 48 in the ECU 30 calculates the CA50 based on the in-cylinder pressure and the crank angle detected in S100. Note that the CA50 in this routine is an actual measurement value.

  Next, it is determined whether or not the ignition FB control can be executed (S104). Specifically, ECU 30 determines whether or not the system configuration of engine 10 operates normally. If it is determined that the FB control cannot be executed, this routine is repeated.

On the other hand, if it is determined in S104 that the ignition FB control can be executed, the annealing process is performed on CA50 and stored as the annealing value CA50 _sm (S106). This process is performed by the annealing process unit 36 in the ECU 30.

The annealing process of S106 is performed by the following equation 2. The SM in the following formula 2 is the number of annealing processes.

Then, the difference value DerutaCA50 _sm the smoothed value CA50 _sm target value _{CA50 tgt} is calculated (S108). The target value CA50 _tgt is set in advance in the target CA50 calculation unit 32 in the ECU 30 as a value for outputting an optimum torque.

Next, the difference value DerutaCA50 _sm, the correction value [Delta] SA _PI based on PI control is calculated (S110). Specifically, PI control unit 40 in the ECU30, from ΔCA50 _sm, P term to calculate the I term, and calculates a sum as the correction value [Delta] SA _PI.

  Next, a subroutine for determining whether or not to execute the next combustion prediction control is executed (S112). This subroutine will be described below with reference to FIG.

[Next combustion prediction control execution possibility determination routine]
FIG. 12 is a flowchart of an execution determination routine for the next combustion prediction control executed by the ECU 30 in the embodiment of the present invention. In the routine shown in FIG. 12, first, CA50 (actual value) is stored (S200). The CA50 stored in S200 is stored as the current measured value until the next CA50 is stored.

Next, the difference value ΔCA50 _D between the previous CA50 and the current CA50 is stored (S202). Specifically, the ECU 30 calculates the difference between the current measured value and the measured value stored immediately before it. Then, ECU 30 is the difference, and stores as a difference value DerutaCA50 _D is a variation of CA50. At this time, the ECU 30 associates the previous actual measurement value (CA50) with the CA50 change amount (ΔCA50 _D ) calculated when the current actual measurement value is measured, and stores it as data.

  Next, it is determined whether or not the number of stored CA50 data is greater than or equal to a predetermined value (S204). When it is determined that the number of stored CA50 data is smaller than the predetermined value, it is determined that the next combustion prediction control cannot be executed (S212). In this case, the ECU 30 determines that the number of data that can be statistically processed is not yet stored. The number of CA50 data stored here refers to the data stored in S202.

On the other hand, in S204, the number of data stored CA50 if it is determined that the predetermined value or more, the correlation coefficient between the CA50 and the difference value ΔCA50 _D (R) is calculated and stored (S206). In this case, the ECU 30 determines that the number of data that can be statistically processed is stored, and statistically processes the stored data.

  Next, it is determined whether or not the correlation coefficient (R) stored in S206 is a predetermined value or more (S208). When it is determined that the correlation coefficient (R) is smaller than the predetermined value, it is determined that the next combustion prediction control cannot be executed (S212).

  On the other hand, when it is determined in S208 that the correlation coefficient (R) is equal to or greater than the predetermined value, it is determined that the next combustion prediction control can be executed (S210). When this determination is made, the ECU 30 determines that a short cycle fluctuation has occurred in the ignition timing. This is the end of the description of the routine for predicting whether or not to execute the next combustion prediction control.

Again, the ignition timing control routine will be described with reference to FIG. The ECU 30 determines whether or not the next combustion prediction control can be executed based on the result of S112 (S114). If the next combustion prediction control is determined to be infeasible in S114, ECU 30 corrects the next ignition timing by the correction value [Delta] SA _PI fraction (S126). The routine is then repeated.

On the other hand, when it is determined in S114 that the next combustion prediction control can be executed, a difference value ΔCA50 between the target value CA50 _tgt and CA50 is calculated (S116).

  Next, a subroutine of a proportional gain determination method is executed (S118). This subroutine will be described below with reference to FIG.

[Proportional gain determination routine]
FIG. 13 is a flowchart of a proportional gain determination routine executed by the ECU 30 in the embodiment of the present invention. In the routine shown in FIG. 13, first, the number of times of combustion (BN) is counted up (S300). The reason for counting the number of combustion is to determine whether or not the stored data is statistically significant.

  Next, it is determined whether the number of combustion times (BN) is equal to or greater than a predetermined number (S302). When it is determined that the number of combustion times (BN) is less than the predetermined number, this routine ends.

  On the other hand, when it is determined that the number of combustion (BN) is equal to or greater than the predetermined number, it is determined whether or not the correlation coefficient (R) is equal to or greater than a predetermined value (S304). If it is determined that the correlation coefficient (R) is smaller than the predetermined value, the proportional gain is subtracted by the predetermined value (S312).

On the other hand, in S304, if the correlation coefficient (R) is determined to be equal to or greater than the predetermined value, the distribution range of data stored in association CA50 and the difference value DerutaCA50 _D respectively (distribution area) (S) is calculated (S306).

Next, it is determined whether or not the current distribution area (S _{current value} ) is smaller than the _previous distribution area (S _{previous value} ) (S308). If it is determined that the current distribution area (S _{current value} ) is greater than or _equal to the previous distribution area (S _{previous value} ), the predetermined value is subtracted from the proportional gain (S312). Next, the combustion frequency (BN) is cleared to zero (S314). Next, this routine ends. The reason why the number of combustions (BN) is cleared to zero is to collect data reflecting the changed proportional gain more than a predetermined number of times and perform statistical processing.

On the other hand, when it is determined that the current distribution area (S _{current value} ) is smaller than the _previous distribution area (S _{previous value} ), a predetermined value is added to the proportional gain (S310). Next, the combustion frequency (BN) is cleared to zero (S314). Next, this routine ends. This is the end of the description of the proportional gain determination routine.

Again, the ignition timing control routine will be described with reference to FIG. S118 following to the relative difference value DerutaCA50, the correction value [Delta] SA _P based on P control is calculated (S120). Specifically, based on the proportional gain set by S118, P control unit 44 in the ECU30 calculates a correction value [Delta] SA _P.

Next, an ignition timing correction value ΔSA is calculated (S122). Specifically, ECU 30 is the difference between the correction value [Delta] SA _P calculated by the correction value [Delta] SA _PI and S120 calculated in S110 is calculated as the correction value [Delta] SA.

  Next, the next ignition timing is corrected by the correction value ΔSA (S124). Then, this routine is repeated. This is the end of the description of the ignition timing control routine.

  In the present embodiment, the CA50 of the combustion ratio is used to correct the ignition timing, but the present invention is not limited to this. For example, the combustion period, that is, the period in which the combustion ratio is within a predetermined range (for example, between 10% and 90%), the crank angle that gives the maximum in-cylinder pressure, or the in-cylinder pressure at the predetermined crank angle, the combustion state correlated with the torque A quantity may be used.

  FIG. 9 shows three types of distribution ranges A, B, and C, but the present invention is not limited to this. The distribution range can be divided into arbitrary types according to the obtained data.

  Here, the ECU 30 executes the above-described S102, so that the “combustion state parameter calculation means” in the first invention performs the above-mentioned S124, and the “ignition timing correction means” in the first invention results from the above-described S124. By executing S106, the “average processing means” in the fourth invention executes the above S126, and the “feedback control means” in the fourth invention executes the above S200 to execute “S200” in the sixth invention. When the “combustion state parameter storage means” executes S202, the “combustion state parameter change amount calculation means” and the “data storage means” according to the sixth invention execute “S206”. The “correlation coefficient calculating means” executes the above S212 to achieve the sixth invention. When the “prohibiting means” performs the above S300, the “measurement means” in the seventh invention executes the above S306, and the “inhibition number determining means” in the seventh invention executes the above S306 by executing the above S302. Thus, the “distributed area calculating means” in the seventh invention realizes the “proportional gain calculating means” in the seventh invention by executing S310 and S312.

10 Engine 16 Crank angle sensor 20 In-cylinder pressure sensor 22 Spark plug 30 ECU
32 Target ignition timing calculation unit 34 Target CA50 calculation unit 36 Smoothing processing unit 40 PI control unit 44 P control unit 48 CA50 calculation unit 50 Next ignition timing calculation unit

Claims (5)

An in-cylinder pressure sensor;
In a control device for an internal combustion engine comprising a crank angle sensor,
Combustion state parameter calculating means for calculating a combustion state parameter whose value is determined for each combustion cycle in accordance with the relationship between the in-cylinder pressure and the crank angle;
Target value setting means for setting a target combustion state parameter that is a target value of the combustion state parameter;
Wherein determining the correction amount of the ignition timing in proportion to the difference between the target combustion state parameter and the combustion state parameter, with respect to the combustion state parameter is the target combustion state parameter indicates a value obtained when the fast combustion and advancing the ignition timing in accordance with the correction amount when are hidden, to the combustion state parameter is the target combustion state parameter, in accordance with the correction amount when the shows the values obtained when the combustion is slow ignition Ignition timing correction means for retarding the timing;
A measuring means for measuring the number of times of combustion in the combustion cycle of each cylinder;
Combustion number determination means for determining whether or not the number of combustion times is a certain number of times;
Combustion state parameter change amount calculating means for calculating a combustion state parameter change amount that is a change amount with respect to the combustion state parameter stored immediately before the combustion state parameter;
Data storage means for storing the combustion state parameter stored immediately before the combustion state parameter and the combustion state parameter change amount as data in association with each other;
A correlation coefficient calculating means for calculating a correlation coefficient between the combustion state parameter stored in the data storage means and the combustion state parameter change amount when the number of data in the data storage means exceeds a certain number;
When the number of combustions is equal to or greater than a certain number and the correlation coefficient is equal to or greater than a certain value, the data storage means has a plane set with the combustion state parameter and the combustion state parameter variation as an axis. A distribution area calculation means for calculating a distribution area of stored data;
When the distribution area calculated by the distribution area calculation means is smaller than the distribution area calculated immediately before, the proportional gain of the correction amount of the ignition timing with respect to the difference between the target combustion state parameter and the combustion state parameter is increased, Proportional gain calculation means for reducing the proportional gain in the above case;
A control device for an internal combustion engine, comprising: The combustion state parameter is a specific combustion rate point that is a crank angle at a specific combustion rate of each cylinder,
The ignition timing correction means corrects the ignition timing to the retard side when the difference between the specific combustion ratio point and a target specific combustion ratio point that is a target value of the specific combustion ratio point is a positive number, and the specific combustion 2. The control apparatus for an internal combustion engine according to claim 1, wherein the ignition timing is corrected to the advance side when the difference between the ratio point and the target specific combustion ratio point is a negative number. Average processing means for performing an average process on the combustion state parameter to obtain an average post-combustion state parameter;
Feedback control means for performing feedback control of ignition timing using a difference between the target combustion state parameter and the average post-combustion state parameter;
The control apparatus according to claim 1 or 2, characterized in that it comprises a. The feedback control unit corrects the ignition timing based on a correction value obtained using any one of P control, PI control, and PID control for a difference between the target combustion state parameter and the average post-combustion state parameter. The control device for an internal combustion engine according to claim 3 . Combustion state parameter storage means for storing the combustion state parameter;
Combustion state parameter change amount calculating means for calculating a combustion state parameter change amount that is a change amount with respect to the combustion state parameter stored immediately before the combustion state parameter;
Data storage means for storing the combustion state parameter stored immediately before the combustion state parameter and the combustion state parameter change amount as data in association with each other;
A correlation coefficient calculating means for calculating a correlation coefficient between the combustion state parameter stored in the data storage means and the combustion state parameter change amount when the number of data in the data storage means exceeds a certain number;
Prohibiting means for prohibiting correction of the ignition timing by the ignition timing correcting means when the correlation coefficient is smaller than a certain value;
The control device for an internal combustion engine according to any one of claims 1 to 4 , further comprising:

Images