JPH09133042A - Cylinder pressure detector device for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To high accurately correct dispersion of a cylinder pressure detection value between cylinders. SOLUTION: In an internal pressure classified by each cylinder detected by prescribed crank angle timing during a compression stroke (S2), an averaged value PAVE is calculated (S3). The internal pressure classified by each cylinder, on which the averaged value PAVE is based, corresponds thereto, to be stored in a map (S4). Similarly, in also a cylinder internal pressure detected by prescribed crank angle timing in a combustion stroke during a fuel cut, correlation between the averaged value PAVE and each detection value is stored in the map. Here, in the case when the internal pressure is changed to increase, the map learned during the compression stroke is referred, and in the case when the internal pressure is changed to decrease, the map learned during the combustion stroke at fuel cut time is referred, and a sensor detection value is converted into the corresponding averaged value PAVE to be output.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an in-cylinder pressure detecting device for an internal combustion engine, and more particularly to a technique for correcting detection variations of in-cylinder pressure detecting means provided for each cylinder.

[0002]

2. Description of the Related Art Conventionally, indicated average effective pressure (or indicated average effective pressure equivalent value) Pi for each cylinder based on the detection result of an in-cylinder pressure sensor (in-cylinder pressure detection means) provided in each cylinder of an engine.
Is calculated and the misfire diagnosis is performed based on the indicated mean effective pressure Pi, or the surge torque is detected based on the fluctuation of the indicated mean effective pressure Pi to control the air-fuel ratio and the ignition timing. It was

[0003]

By the way, in the in-cylinder pressure sensor, there is an output variation in the sensor itself, and
Especially, a ring-shaped piezoelectric element is attached as a washer for the spark plug,
When a sensor configured to detect the in-cylinder pressure as a relative pressure to the tightening load of the spark plug is used, the sensor output varies due to the variation of the tightening load in addition to the variation of the sensor itself.

Therefore, for example, even when there is actually no difference in the indicated mean effective pressure between the cylinders, an apparent deviation in the indicated mean effective pressure between the cylinders occurs due to the variation in the sensor output, and the surge torque There was a fear that would be falsely detected. Further, since the in-cylinder pressure sensor mounted as the washer of the spark plug has a characteristic of having hysteresis on the pressure increasing side and the pressure reducing side (see FIG. 8), the variation of the sensor output is learned with high accuracy. There is a problem that it is difficult to obtain the original in-cylinder pressure on both the pressure increasing side and the pressure reducing side.

The present invention has been made in view of the above problems, and an object of the present invention is to provide an in-cylinder pressure detecting device capable of accurately correcting an output variation of an in-cylinder pressure detecting means provided for each cylinder.

[0006]

Therefore, the invention according to claim 1 is constructed as shown in FIG. In FIG.
In-cylinder pressure detection means is provided in each cylinder of the engine,
The in-cylinder pressure of each cylinder is individually detected. The average value calculation means is
An average value of the in-cylinder pressure for each cylinder detected by the in-cylinder pressure detecting means at a predetermined crank angle timing in the non-combustion state is calculated.

Here, the variation learning means learns, for each cylinder, the variation detected by the in-cylinder pressure detection means with respect to the average value based on the average value calculated by the average value calculation means. Then, the correcting means corrects the detection result for each cylinder by the in-cylinder pressure detecting means based on the learning result by the variation learning means, and finally outputs the correction result as an in-cylinder pressure detection value.

According to this structure, in the non-combustion state, assuming that there is no variation in the charging efficiency among the cylinders, the in-cylinder pressure detected at the same crank angle timing should be the same. The variation of the detection value at indicates the variation of the output of the detecting means. Therefore, assuming that the average value of the in-cylinder pressure detected at the same crank angle timing is used as a reference value, and the same detection result as the reference value should have been output from each in-cylinder pressure detecting means, Variation learning is performed separately.
Then, the in-cylinder pressure is corrected for each cylinder on the basis of the learning result, and the in-cylinder pressure deviation due to the sensor variation, which is not the deviation due to the combustion influence, is avoided between the cylinders.

According to the second aspect of the invention, the average value calculating means calculates the average value at a predetermined crank angle timing during the compression stroke before the compression top dead center.
With this configuration, it is possible to learn the variation in the in-cylinder pressure detection means when the in-cylinder pressure increases and changes. In the invention according to claim 3, the average value calculating means calculates the average value at a predetermined crank angle timing in the combustion stroke after the compression top dead center during the fuel cut.

According to such a configuration, it is possible to learn the variation of the in-cylinder pressure detecting means when the in-cylinder pressure decreases and changes at a predetermined crank angle timing that is normally affected by combustion. In the invention according to claim 4, the variation learning means separately performs learning when the in-cylinder pressure increases and changes and learning when the in-cylinder pressure decreases, and the correction means detects the in-cylinder pressure detection means. The learning result by the variation learning means is selectively used for correction depending on whether the in-cylinder pressure is increased or decreased.

According to this structure, the in-cylinder pressure detecting means is
Even when the output characteristic has a hysteresis when the in-cylinder pressure increases and decreases, even if the variation learning can be performed with high accuracy, the learning result is selectively used based on the direction of the in-cylinder pressure change during correction. Highly accurate correction is possible. According to the invention of claim 5, the variation learning means is configured to estimate and learn variation by interpolation through a region in which the maximum in-cylinder pressure during non-combustion is exceeded.

According to this structure, it is possible to estimate the variation on the high pressure side, which cannot be directly learned, and correct the detected pressure during combustion in accordance with the variation detected by the in-cylinder pressure detection means.

[0013]

Embodiments of the present invention will be described below. In FIG. 2 showing the system configuration, air is taken into the internal combustion engine 1 through an air cleaner 2, an intake duct 3, and an intake manifold 4. In the intake duct 3,
A butterfly-type throttle valve 5 that interlocks with an accelerator pedal (not shown) is interposed, and the throttle valve 5 adjusts the intake air amount of the engine.

An electromagnetic fuel injection valve 6 is provided for each cylinder in each branch portion of the intake manifold 4, and a predetermined space is provided by electronically controlling the amount of fuel injected and supplied from the fuel injection valve 6. A fuel-air mixture is formed. Here, by individually controlling the fuel injection valve 6 provided for each cylinder, it is possible to form an air-fuel mixture having a different air-fuel ratio for each cylinder.

The air-fuel mixture sucked into the cylinder through the intake valve 7 is ignited and burned by spark ignition by the spark plug 8 provided for each cylinder, and the combustion exhaust gas is discharged through the exhaust valve 9 and the exhaust manifold. Catalyst not shown by 10,
Guided by the muffler. The control unit 11 for controlling the fuel injection amount by the fuel injection valve 6 and the ignition timing of the spark plug 8 is constituted by including a microcomputer, and the intake air amount signal Q from the hot wire type air flow meter 12 and the throttle sensor 13 The throttle valve opening signal TVO, the crank angle signal from the crank angle sensor 14, the cooling water temperature signal Tw from the water temperature sensor 15, the cylinder pressure signal P from the cylinder pressure sensor 16, and the like are input.

The hot-wire type air flow meter 12 directly detects the intake air amount of the engine 1 as a mass flow rate based on the resistance change of the temperature-sensitive resistance due to the intake air amount.
The throttle sensor 13 is an opening TV of the throttle valve 5.
O is detected by a potentiometer. The crank angle sensor 14 outputs a unit angle signal for each unit crank angle and a reference angle signal for each predetermined piston position. Here, the engine rotation speed Ne can be calculated by measuring the number of generations of the unit angle signal within a predetermined time or the generation cycle of the reference angle signal.

The water temperature sensor 15 detects the cooling water temperature Tw in the water jacket of the engine 1.
The in-cylinder pressure sensor 16 (in-cylinder pressure detection means) is an actual open cylinder 63
A sensor that detects the combustion pressure as a relative pressure with respect to the tightening load of the ignition plug, which is composed of a ring-shaped piezoelectric element mounted as a washer of the ignition plug 8 as disclosed in Japanese Patent Publication No. 17432. By mounting the spark plug 8 of each cylinder, the cylinder pressure P (combustion pressure) can be detected for each cylinder. The in-cylinder pressure sensor 16 is of a type that is mounted as a washer of the spark plug 8 as described above, and is of a type that directly detects the in-cylinder pressure in the combustion chamber and detects the in-cylinder pressure as an absolute pressure. Is also good.

The control unit 11 determines the basic ignition timing (basic ignition advance value) based on engine operating conditions such as engine load and engine rotation speed, and controls the ignition timing by the spark plug 8. The control of the injection amount of the fuel injection valve 6 by the control unit 11 is performed as follows.

Based on the intake air amount Q detected by the hot wire air flow meter 12 and the engine speed Ne calculated from the detection signal from the crank angle sensor 14, the basic fuel injection amount Tp (corresponding to the target air-fuel ratio = K × Q / Ne: K is a constant), and the cooling water temperature T is added to the basic fuel injection amount Tp.
A final fuel injection amount Ti is obtained by performing a correction according to an operating condition such as w. Then, a drive pulse signal having a pulse width corresponding to the fuel injection amount Ti is output to the fuel injection valve 6 at a predetermined timing. The fuel, which is adjusted to a predetermined pressure by a pressure regulator (not shown), is supplied to the fuel injection valve 6, and an amount of fuel proportional to the pulse width of the drive pulse signal is injected and supplied to a predetermined space. A fuel-air mixture is formed.

Further, the control unit 11 detects the combustion state based on the detection signal from the in-cylinder pressure sensor 16 and performs the correction control of the fuel injection amount and the ignition timing based on the detection result. The in-cylinder pressure sensor 16 has variations in output due to variations in the sensor itself and variations in tightening load, and there is a risk of erroneous detection of variations in combustion pressure (surge torque), especially between cylinders. Therefore, as shown in the flowcharts of FIGS. 3 to 5, the output variation is learned, and the detection value is corrected based on the learning result.

The control unit 11 has a function as an average value calculation means, a variation learning means, and a correction means, as shown in the flow charts of FIGS. In the flowchart of FIG. 3, in step 1 (denoted as S1 in the figure. The same applies hereinafter), each cylinder pressure sensor
The in-cylinder pressures P # 1 to P # 4 (for four cylinders) for each cylinder detected in 16 are sampled.

In step 2, it is judged whether or not it is a predetermined crank angle timing which is set in advance during the compression stroke before the compression top dead center. It should be noted that, as shown in FIG. 6, a plurality of the predetermined crank angle timings are set at regular intervals during the compression stroke. When it is determined in step 2 that the predetermined crank angle timing is reached, the process proceeds to step 3,
Calculating the same crank each cylinder of the cylinder pressure sensing value sampled by the retarded timing P # 1 to P # average P _AVE of 4.

P_AVE= (P # 1 + P # 2 + P # 3 + P # 4) / 4 In the next step 4, the average value P_AVEEach cylinder that became the basis of
Cylinder pressure of the average value P _AVEIt is made to correspond and is memorized. Immediately
Then, from the detection result of the cylinder pressure sensor 16, the average value P_AVESee
The average value P_AVEInside the cylinder
Since the pressure is the value sampled in the non-combustion state,
The in-cylinder pressure sensor 16 is not affected by the variation in combustion in each cylinder.
It will vary depending on the output variation. Therefore, see
Illuminated average value P_AVEThe deviation between the detected value and
The average value P indicating the output variation and the detection result_AVE
If replaced with, the output variation of the in-cylinder pressure sensor 16 is eliminated.
The in-cylinder pressure data thus obtained can be obtained. Paraphrase
For example, unify the characteristics of the detection output with respect to the cylinder pressure to the average value.
Will be corrected.

Further, by making the variation learning limited to the crank angle timing during the compression stroke, it is possible to learn only the variation characteristic when the in-cylinder pressure increases, and output in the increasing and decreasing directions of the in-cylinder pressure. It is possible to prevent the accuracy of the variation learning from deteriorating due to the hysteresis when there is hysteresis. The flowchart of FIG.
This is for learning the output variation when the in-cylinder pressure decreases and changes. In step 12, it is judged whether or not it is a predetermined crank angle timing in the combustion stroke after the compression top dead center during fuel cut (see FIG. 7). ), The difference from the flowchart of FIG. 3 is that only the average value P _AVE is calculated in the combustion stroke during fuel cut.

Here, each step 11 other than step 12,
For 13 and 14, S in the flowchart of FIG.
Since the same processing as 1, 3, 4 is performed, the description will be omitted.
The condition of the non-combustion state during the fuel cut is to eliminate the influence of the combustion variation between the cylinders and to learn only the output variation of the in-cylinder pressure sensor 16. The characteristic of the output variation of the in-cylinder pressure sensor 16 can be learned.

The flow chart of FIG. 5 is as described above.
It shows how to correct the in-cylinder pressure detection value based on the result of the variation learning which is performed separately when the in-cylinder pressure increases and when the in-cylinder pressure changes in the non-combustion state. Step 21
Then, the detection result by the in-cylinder pressure sensor 16 is read, and in step 22, it is determined whether the changing direction of the in-cylinder pressure at the time of detecting the in-cylinder pressure is the increasing direction or the decreasing direction.

Then, for example, when there is an increase change, the routine proceeds to step 23, where the relative relationship between the average value P _AVE learned based on the flowchart of FIG. The corresponding average value _AVE is obtained, and the average value _AVE is output as the final detection result.
On the other hand, if there is a decrease change, the routine proceeds to step 24, and the relative value between the average value P _AVE learned based on the flowchart of FIG. 4 and the detection result by each sensor is referred to, and the average value corresponding to the detection result. _AVE is obtained, and the average value _AVE is output as the final detection result.

Since the final detection result of the in-cylinder pressure is a value excluding the influence of the output variation of the in-cylinder pressure sensor 16, for example, the in-cylinder pressure or the indicated mean effective pressure based on the in-cylinder pressure is compared between the cylinders. It is possible to prevent erroneous detection of sensor variation as surge torque when detecting the occurrence of surge torque. Note that the variation learning is performed only during non-combustion and it is not possible to directly learn the pressure exceeding the maximum pressure during non-combustion. It is preferable to learn the variation characteristic in.

Further, in the above description, the relative relationship between the average value P _AVE and the detection result of each sensor is stored in the map, but the detection result of each sensor is shifted by a certain amount with respect to the average value P _AVE . If it has, the shift amount may be learned for each sensor, or the average value P _AVE may be learned.
You may make it learn the variation of each sensor as a ratio with respect to. However, in this case as well, it is preferable to perform learning and correction separately when the in-cylinder pressure increases and when it decreases.

[0030]

As described above, according to the first aspect of the present invention, it is possible to correct the variation of the detection result for each cylinder due to the variation of the in-cylinder pressure detecting means, and the variation of the sensor which is not the variation due to the combustion influence between the cylinders. There is an effect that it is possible to prevent the cylinder pressure deviation from being erroneously detected.

According to the second aspect of the invention, there is an effect that the variation of the in-cylinder pressure detecting means can be learned when the in-cylinder pressure increases and changes in the compression stroke. According to the third aspect of the present invention, it is possible to learn the variation of the in-cylinder pressure detecting means when the in-cylinder pressure decreases and changes at a predetermined crank angle timing during a combustion stroke that is normally affected by combustion.

According to the fourth aspect of the invention, even if the output characteristic of the in-cylinder pressure detecting means has a hysteresis when the in-cylinder pressure increases and when the in-cylinder pressure changes, a variation learning can be performed with high accuracy and a high level. There is an effect that correction can be performed with high accuracy. According to the invention described in claim 5, there is an effect that it is possible to estimate the variation on the high pressure side that occurs during combustion, which cannot be directly learned in the non-combustion state, and correct the detection variation of the in-cylinder pressure detection means in all regions.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an apparatus according to the invention of claim 1.

FIG. 2 is a system configuration diagram of the engine in the embodiment.

FIG. 3 is a flowchart showing variation learning when the in-cylinder pressure increases.

FIG. 4 is a flowchart showing variation learning when the in-cylinder pressure decreases.

FIG. 5 is a flowchart showing variation correction control.

FIG. 6 is a diagram showing a cylinder pressure sampling timing during a compression stroke.

FIG. 7 is a diagram showing in-cylinder pressure sampling timing during a combustion stroke during fuel cut.

FIG. 8 is a diagram showing a hysteresis characteristic of an in-cylinder pressure sensor.

[Explanation of symbols]

 1 Internal Combustion Engine 4 Intake Manifold 5 Throttle Valve 6 Fuel Injection Valve 8 Spark Plug 10 Exhaust Manifold 11 Control Unit 12 Hot Wire Air Flow Meter 13 Throttle Sensor 14 Crank Angle Sensor 15 Water Temperature Sensor 16 Cylinder Pressure Sensor

Claims (5)

[Claims] 1. An in-cylinder pressure detecting means provided in each cylinder of an engine for individually detecting the in-cylinder pressure of each cylinder, and each in-cylinder pressure detecting means at a predetermined crank angle timing in a non-combustion state. Based on the average value calculating means for calculating the average value of the in-cylinder pressure for each cylinder, and based on the average value calculated by the average value calculating means, the variation in detection by the in-cylinder pressure detecting means with respect to the average value is learned for each cylinder. Variation learning means, and correction means for correcting the detection result for each cylinder by the in-cylinder pressure detection means based on the learning result by the variation learning means, and finally outputting the correction result as an in-cylinder pressure detection value. An in-cylinder pressure detection device for an internal combustion engine, comprising: 2. An in-cylinder pressure detecting device for an internal combustion engine according to claim 1, wherein said average value calculating means calculates the average value at a predetermined crank angle timing during a compression stroke before compression top dead center. 3. The in-cylinder pressure of the internal combustion engine according to claim 1, wherein the average value calculating means calculates the average value at a predetermined crank angle timing in a combustion stroke after compression top dead center during fuel cut. Detection device. 4. The cylinder for which the variation learning means individually performs learning when the in-cylinder pressure increases and learns when the in-cylinder pressure decreases, and the correcting means detects the cylinder pressure by the in-cylinder pressure detecting means. The in-cylinder pressure detection device for an internal combustion engine according to any one of claims 1 to 3, wherein the learning result by the variation learning unit is used for correction depending on whether the internal pressure is increasing or decreasing. 5. The internal combustion engine according to any one of claims 1 to 4, wherein the variation learning means estimates and learns a variation in a region exceeding the maximum in-cylinder pressure in a non-combustion state by interpolation. In-cylinder pressure detector.