JP2003013793A - Internal combustion engine misfire detection method and misfire detection system - Google Patents

Abstract

(57) [Problem] To provide a misfire determination method for an internal combustion engine that can be performed simply and inexpensively by using a pressure seat type pressure sensor, and that can perform misfire determination with high accuracy and good reproducibility. SOLUTION: A pressure sensor (seat type pressure sensor) provided on a mounting seat of a spark plug measures an in-cylinder pressure of an internal combustion engine to which the spark plug is mounted. The first pressure measurement time is determined in the period before the top dead center until the crank angle reaches the top dead center after the intake valve closes, and the measured value of the in-cylinder pressure at the first pressure measurement time is P1, and the crank angle is A second pressure measurement time is determined in a period after the top dead center from when the top dead center is reached until the exhaust valve opens, and a measured value of the in-cylinder pressure at the second pressure measurement time is P2. In addition, a correction measurement time is set within the period before the top dead center and before the ignition timing, and the in-cylinder pressure measurement value at the correction measurement time is set as a correction pressure value P0, and the correction pressure value is calculated from the pressure measurement values P1 and P2. Values obtained by subtracting P0 are defined as P1 'and P2'. Then, a misfire determination of the internal combustion engine is performed based on the ratio between P1 'and P2'.

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a misfire determination method and system for an internal combustion engine.

[0002]

2. Description of the Related Art As a misfire determination method in an internal combustion engine, for example, a pressure guide hole communicating with a combustion chamber is formed in a cylinder head, and a partition type pressure sensor is arranged therein to directly measure the pressure in the combustion chamber. A method of measuring can be considered. However, a process for forming the pressure guide hole in the cylinder head is necessary, and it is inevitable that the structure becomes complicated and the manufacturing cost increases. On the other hand, as means for more simply measuring the in-cylinder pressure, for example, Japanese Patent Laid-Open No. 6-290853 discloses
A method for measuring the in-cylinder pressure by a pressure sensor (hereinafter referred to as a seat pressure sensor) provided on a mounting seat of a spark plug is disclosed.

FIG. 3 illustrates an in-cylinder pressure measurement profile for one cycle in the combustion chamber, which is measured by an in-cylinder pressure sensor. The solid line shows the profile during normal combustion, and the one-dot chain line shows the profile during misfire. It is shown. When the intake valve is closed, the inside of the combustion chamber is closed, and the rise of the piston compresses the air-fuel mixture to increase the in-cylinder pressure. And the top dead center (TDC)
r) The spark plug is ignited at a certain crank angle (ignition timing) before. When the ignition of the air-fuel mixture is normally performed by this ignition, the cylinder pressure is further increased by the explosion of the air-fuel mixture. The pressure increment causes the piston to move to TDC
The cylinder pressure continues to rise for a while even after passing through the point and starts decreasing, and then decreases when the piston moves down to some extent. Therefore, the in-cylinder pressure measurement profile is an asymmetric profile in which the peak position is shifted to a higher angle side than the crank angle (αTDC) corresponding to TDC. On the other hand, for example, if the ignition ends abnormally and there is a misfire, there is no pressure increase due to the explosion, so the pressure becomes maximum at TDC where the volume of the combustion chamber is the minimum, and then the pressure change that is just opposite to that when rising is followed. The piston moves down. Therefore, the in-cylinder pressure measurement profile has a substantially symmetrical peak position that coincides with αTDC. As described above, there is a difference in the peak value of the in-cylinder pressure measurement profile between normal combustion and misfire, and a clear difference in profile symmetry.

However, the misfire determination using only one in-cylinder pressure peak value level has many error factors such as failure to detect misfire when complete misfire does not occur, and there is a drawback in lack of accuracy. Therefore, in each of JP-A-57-80538, JP-A-61-23876 and JP-A-1-172730, the in-cylinder pressure measurement value at the first pressure measurement time set within the period before top dead center is disclosed. Is set to P1, and the in-cylinder pressure measurement value at the second pressure measurement time determined within the period after top dead center is set to P2.
There is disclosed a method of making a misfire determination based on the ratio of P2 to P2.

[0005]

The method using a seat pressure sensor utilizes the fact that the tightening force of the spark plug fixing the sensor is relaxed by receiving the cylinder pressure, so to speak indirectly measures the cylinder pressure. To do. Therefore, compared to the method of directly measuring with a partition type sensor through the pressure guide hole, there are many factors that cause fluctuations in the pressure measurement value, such as the tightening force of the spark plug and variations in piezoelectric element performance. Accurate misfire detection using absolute values is virtually impossible.

An object of the present invention is to provide a misfire determination method and system for an internal combustion engine, which can be easily and inexpensively implemented by using a seat-type pressure sensor and which is capable of highly accurate and reproducible misfire determination. To do.

[0007]

In order to solve the above-mentioned problems, a method for judging a misfire of an internal combustion engine according to the present invention is based on the output of a pressure sensor provided on a mounting seat of the spark plug, Cylinder pressure measurement based on the cylinder pressure of the internal combustion engine to which is installed is acquired, and the first pressure measurement is performed within the period from the closing of the intake valve until the crank angle reaches top dead center (pre-top dead center period). The time is determined, the cylinder pressure measurement value at the first pressure measurement time is set to P1, and the second pressure measurement is performed within the period (after the top dead center) from when the crank angle reaches the top dead center until the exhaust valve opens. Set the time,
The in-cylinder pressure measurement value at the second pressure measurement time is set to P2, and on the other hand, the correction measurement time is set within the period before the top dead center and before the ignition timing, and the in-cylinder pressure measurement value at the correction measurement time is set to the correction pressure. A value P0 is set, and values obtained by subtracting the corrected pressure value P0 from the measured pressure values P1 and P2 are set as P1 'and P2', and misfire determination of the internal combustion engine is performed based on the ratio of P1 'and P2'. And

Further, the misfire determination system for an internal combustion engine according to the present invention includes a pressure sensor provided on a mounting seat of the spark plug in order to obtain a measured value of the in-cylinder pressure based on the in-cylinder pressure of the internal combustion engine to which the spark plug is attached. Then, the first pressure measurement time is set within the period from the closing of the intake valve until the crank angle reaches the top dead center (the period before top dead center), and the cylinder pressure measurement value at the first pressure measurement time is defined as P1. , The second pressure measurement time is set within the period (after the top dead center) from when the crank angle reaches the top dead center until the exhaust valve opens,
The in-cylinder pressure measurement value at the second pressure measurement time is set to P2, and on the other hand, the correction measurement time is set within the period before the top dead center and before the ignition timing, and the in-cylinder pressure measurement value at the correction measurement time is set to the correction pressure. A value P0, a value obtained by subtracting the corrected pressure value P0 from the pressure measurement values P1 and P2, respectively, and P1 'and P2', and a judgment unit for judging the misfire of the internal combustion engine based on the ratio of P1 'and P2'. It is characterized by having.

The measurement time may be measured by timer measurement with the time t as a parameter, or the crank angle α may be measured and the measured value of α may be used as the time measurement value.

In the misfire determination method and system of the present invention, the first pressure measurement time is set within the period before top dead center, and the cylinder pressure measurement value at the first pressure measurement time is P1.
Then, the second pressure measurement time is determined within the period after top dead center, and the cylinder pressure measurement value at the second pressure measurement time is P2.
Then, basically, the misfire determination of the internal combustion engine is performed based on the pressure measurement values P1 and P2. When the in-cylinder pressure measurement value fluctuates due to the above factors, the in-cylinder pressure measurement profile showing the time variation of the in-cylinder pressure measurement value is shown in FIG.
As shown in (a) and (b), the peak height changes. However, the difference in waveform symmetry between misfire and normal does not change even if the peak height changes a little. Therefore, pressure measurement times are set in the pre-dead-center period and the post-dead-center period, respectively, and the in-cylinder pressure measurement values P1 and P of the two are determined.
The presence or absence of misfire can be determined by referring to the ratio with 2.

However, in the misfire determination using the seat-type pressure sensor, as described above, the measured value of the in-cylinder pressure is affected by the tightening force of the spark plug on the sensor, the variation of the piezoelectric element performance, the pyroelectric characteristics, and the like. Easy to use. Therefore, in the present invention, a correction method for standardizing the measured value of the in-cylinder pressure by comparison with a reference value is found, and it is possible to suppress the fluctuation of the measured value of the in-cylinder pressure due to the factors as described above, which in turn leads to a higher value. Accurate misfire determination is possible. The correction measurement time is set within the period before top dead center which is relatively insensitive to the pressure increase due to the ignition / explosion of the fuel, and the cylinder pressure measurement value at the correction measurement time is set as the correction pressure value P0. And P2, the values obtained by subtracting the corrected pressure value P0 from each other are defined as P1 'and P2'.
The misfire determination is performed based on the ratio of 1'and P2 '.

That is, in the seat type pressure sensor used for measuring the in-cylinder pressure, the sensor element is made of piezoelectric ceramic. On the other hand, since many piezoelectric ceramics show pyroelectricity due to temperature rise, the pressure sensor element using this has a problem that the zero point of the sensor output easily drifts due to temperature change. FIG. 7A schematically shows the in-cylinder pressure measurement profile in the steady state where the temperature change is small, and FIG. 7B schematically shows the in-cylinder pressure measurement profile in the transient state where the temperature change is rapid. The effect of zero-point drift due to temperature is such that the in-cylinder pressure measurement profile is uniformly shifted over the entire measurement period in a short period in which the temperature change does not pose a problem.
Therefore, by displaying the in-cylinder pressure measurement values P1 and P2 by increments P1 ′ and P2 ′ from the reference pressure measurement value P0, the influence of the zero point drift can be reduced.

It is more desirable to set the corrected measurement time to a time after the intake valve is closed. At the time of opening / closing the intake valve or the exhaust valve, due to the influence of the distortion of the cylinder head, a corresponding distortion may occur in the output of the seat-type pressure sensor as shown by the circled portion in FIG. . However, if the in-cylinder pressure measurement value is acquired after the intake valve is closed, it is not necessary to be affected by the distortion.

Further, since the in-cylinder pressure measurement profile at the time of misfire is ideally symmetrical, as shown in FIG. 5, the first pressure measurement time αJ and the second pressure measurement time αK are at the top dead center. It can be said that the clearest and easiest method is to set the times at which the crank angles from are equal to each other. This is because if the profile at the time of misfire (shown by the one-dot chain line in the figure) is completely symmetrical, as is apparent from FIG. 5 and the like, the measured values P1 and αK at αJ
This is because the measured value P2 in 1 becomes equal to the measured value P2. in this case,
If P1 and P2 are strictly equal (for example, P2 / P1 =
If it is 1) misfire, if it is slightly different (eg P
If it is 2 / P1> 1), it is theoretically possible to judge that it is normal. However, as a practical problem, due to various error factors (for example, the influence of hysteresis described later), P1 does not become P2 even at the time of misfire, for example, P2> P.
Sometimes measured as 1. In this case, in consideration of this error amount, a slight margin is provided in the determination standard of P2 / P1, and if P2 / P1 becomes smaller than a certain lower limit value that is larger than 1, it is determined that a misfire has occurred. Is more effective in avoiding erroneous determination. Although it is a slightly odd method, the first pressure measurement time αJ and the second pressure measurement time αK are set asymmetrically with respect to the top dead center αTDC by an amount corresponding to the margin of the above-mentioned criterion.
For example, when P2 / P1 becomes 1 or less, misfire can be determined to be normal if it is larger than 1. In such a case, the deviations of αJ and αK from αTDC are not exactly equal.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an example of a mounting form of a seat type pressure sensor. That is, the spark plug 50 has the mounting screw portion 52 formed on the outer peripheral surface of the tip end portion of the metal shell 51, and the spark discharge gap g is smaller than that of the cylinder head SH of the internal combustion engine configured as an automobile engine. The mounting screw portion 52 is mounted by being screwed into a screw hole TH formed in the bottom of the plug hole PH so as to be located in the combustion chamber ER. Also,
A flanged mounting seat 51f is formed on the outer peripheral surface of the metallic shell 51 so as to be adjacent to the base end of the mounting screw 52. On the other hand, the seat-type pressure sensor 5 has a ring-shaped sensor element 10 made of piezoelectric ceramic, which together with the ring-shaped gasket GS, the mounting seat 51f and the screw hole T.
Between the opening peripheral edge MP of H, a constant bias pressure is applied in the axial direction in a sandwiched manner. Although FIG. 1 shows an embodiment in which the sensor element is integrally incorporated in the spark plug, a sensor element configured separately from the spark plug is used when mounting the spark plug.
You may make it fit in the attachment seat part and attach.

The spark plug 50 receives the pressure in the axial direction due to the generation of the pressure in the combustion chamber ER, that is, the in-cylinder pressure,
The bias pressure applied to the sensor element 10 changes. As a result, the amount of piezoelectric charge generated in the sensor element 10 changes, and the change signal is extracted as in-cylinder pressure measurement information via the output cable CB.

FIG. 2 is a block diagram showing an example of the misfire determination unit of the present invention using the seat type pressure sensor 5. The main body of the misfire determination unit 1 is an ECU 2 composed of a computer, which collectively controls the operation of the internal combustion engine such as the ignition timing and the air-fuel ratio together with the misfire determination. In addition, except for the misfire determination function described below,
Since the configuration and function of the ECU 2 are well known, only the misfire determination function will be described.

The ECU 2 includes a CPU 3, a ROM 4 and a RAM
5 and the input / output interface 6 are configured as a computer connected to the bus. The ROM4 has an ECU
A control program describing the control processing contents of No. 2 is stored, and a misfire determination routine is also incorporated therein. The CPU 3 realizes the function of the ECU 2 by using the RAM 5 as a work memory and executing the control program.

A crank angle sensor 7 for detecting the crank angle of the internal combustion engine is connected to the input / output interface 6. The crank angle sensor 7 is, for example, a pulse generator that detects a rotation angle of a crank shaft,
The pulse signal is input to a predetermined port of the input / output interface 6 via the Schmitt trigger 8. Also,
By the input pulse interval from the crank angle sensor 7, it is possible to monitor the engine speed, and thus the length of one cycle, in real time.

Next, the pressure sensor element 10 is connected to the input / output interface 6 via the charge amplifier circuit 11. In the charge amplifier circuit 11, the output cable from the pressure sensor element 10 is connected to the negative input terminal of the operational amplifier 15, and the positive input terminal is grounded (that is, an inverting amplifier having the ground level as a reference voltage is formed. ing). When pressure is generated in the combustion chamber, electric charges are generated in the pressure sensor element 10 (FIG. 2). As a result, the negative feedback capacitor 13 connected to the operational amplifier 15 accumulates a charge that balances the electric charge, and the terminal voltage thereof is input to the negative input terminal of the operational amplifier 15 as a voltage-converted charge signal. Therefore, the operational amplifier 15, together with the negative feedback capacitor 13, constitutes a charge-voltage conversion circuit that outputs the charge generated in the pressure sensor element 10 as an amplified voltage signal. The resistor 12 inserted in parallel with the negative feedback capacitor 13 promotes the discharge of the negative feedback capacitor 13 when the charge level generated by the pressure sensor element 10 starts to decrease, and thus the operational amplifier 15.
Plays a role in preventing output saturation. The resistor 14 provided on the signal line from the pressure sensor element 10 is for protecting the terminal of the operational amplifier 15. Charge amplifier circuit 1
The output of 1 is input to a predetermined port of the input / output interface 6 via the A / D converter 17 as a cylinder pressure measurement value signal (cylinder pressure measurement information).

The flow of the misfire determination process will be described below with reference to the explanatory diagram of FIG. 4 and the flowchart of FIG. 6 (the memory of each variable used in this process is as shown in the RAM 5 of FIG. 2). Has been formed). First, the crank angle α signal is input as a pulse signal output from the crank angle sensor 7 in FIG. 2, and is added to the α counter in the RAM 5. The added value of the α counter indicates the current crank angle (however, the crank angle sensor 7 can be an absolute pulse generator, and in this case, the α counter is unnecessary). Then, in S1 of FIG. 6, as the initialization processing, the α counter is reset and each memory in the RAM 5 is reset. After the reset, the addition of the α counter is started according to the execution of the cycle start job by the ECU 2.

After that, a predetermined cylinder pressure measurement value P
Of the sampling timing (that is, while continuing the addition of the α counter until the first pressure measurement time, the second pressure measurement time, and the correction measurement time described above arrive. In other words, in S3, the in-cylinder pressure measurement signal and its integration are integrated. The reading of the input port of the value is set to the standby state, the α counter is read, and in S4, the crank angle α indicated by the α counter is the constant crank angle position α1 after the intake valve is closed and before the ignition timing. It is determined whether or not (correction measurement time) has been reached.If No, the process returns to S3 to repeat the following processes, and if Yes, the process proceeds to S5 and thereafter.

In S5, the in-cylinder pressure measurement value P is read in response to the arrival of the correction measurement time α1 and is set in the memory as the value of P0. In S6, sampling standby is again performed, and in S7, it is determined whether or not α has reached the first pressure measurement time αJ. If No, the process returns to S6 to repeat the following processes, and if Yes, the process proceeds to S8 and subsequent processes.

At S8, the in-cylinder pressure measurement value P is read in response to the arrival of the first pressure measurement time αJ (where αTDC-αJ = αK-αTDC), and this is read as P1.
Set in memory as the value of. In S9, sampling standby is again performed, and in S10, it is determined whether or not α has reached the second pressure measurement time αK. If No, the process returns to S9 and repeats the following processes, and if Yes, the process proceeds to S11 and subsequent processes.

In S11, the cylinder pressure measurement value P is read in response to the arrival of the second pressure measurement time αK (where αTDC-αJ = αK-αTDC), and this is read in P2.
Set in memory as the value of. Then, in S12, the measured P0 is subtracted from the values of P1 and P2 to obtain P0.
Calculate the values of 1'and P2 '. Furthermore, in S13,
The value of the judgment index γ is calculated as P2 ′ / P1 ′. The corrected measurement time α1 is set to 90 ° before top dead center here, but it is not limited to this as long as it is before ignition. Then, in S14, the value of γ is compared with a determination reference value (here, an upper limit value) γc, and if γ <γc, it is determined that a misfire has occurred. The ECU 2 of FIG. 2 outputs a predetermined misfire determination output from the determination output port of the input / output interface 6 (FIG. 6: S15). In S16, it is determined whether or not the engine is stopped. If not, the process returns to S1 and the subsequent processes are repeated.

The results of experiments conducted to confirm the effects of the present invention will be described below. First, 10 seat-type pressure sensors with the same specifications as shown in FIG. 1 were prepared and mounted on a gasoline engine with four cylinders and a displacement of 2000 cc together with a spark plug, and the ignition timing was set to before BDC (BTD).
C) The engine was operated at various engine speeds at 15 °, and misfire was judged by the judgment unit in FIG. The first pressure measurement time (cylinder pressure measurement value: P1) is set to 50 before top dead center.
°, the second pressure measurement time (cylinder pressure measurement value: P2) is set to 50 ° after top dead center, and the corrected measurement time (cylinder pressure measurement value: P0) is set to 90 ° before top dead center. Then, the determination is P2.
/ P1 was used, and P2 '/ P1' with the above-described correction for reducing P0 was used. FIG. 8 shows the results. The black rhombus plot points indicate the pressure ratio of the cycle determined to be normal combustion, and the black square plot points indicate the pressure ratio of the cycle determined to be misfire. . The solid error bar is P
The range of variation in the value when 2 '/ P1' is used, the error bar in the broken line indicates the range of variation in the value when P2 / P1 is used, and the plotted points indicate each average value. Further, FIG. 9 summarizes the variation range ratio with respect to the average value for each rotation speed in the above two cases. According to this, it is understood that the variation in the case of using P2 / P1 'at each rotational speed is smaller than that in the case of using P2 / P1, and the determination accuracy is good.

[Brief description of drawings]

FIG. 1 is a vertical cross-sectional view showing an example of a mounting form of a seat-type pressure sensor.

FIG. 2 is a block diagram showing an example of the electrical configuration of the misfire determination unit of the present invention.

FIG. 3 is a diagram showing a comparison of in-cylinder pressure measurement profiles during misfire and normal combustion.

FIG. 4 is a diagram showing the relationship between the in-cylinder pressure measurement profile and the pressure measurement time in FIG.

FIG. 5 is a diagram for explaining how the in-cylinder pressure measurement value level fluctuates.

FIG. 6 is a flowchart showing a flow of misfire determination processing in the misfire determination unit of FIG.

FIG. 7 is a diagram illustrating the effect of temperature drift on the in-cylinder pressure measurement profile.

FIG. 8 is a graph showing the results of an experiment conducted to confirm the effect of the present invention.

FIG. 9 is a graph for explaining how the influence of variations among sensors is reduced by correction.

FIG. 10 is an explanatory diagram showing a state in which the output of the seat-type pressure sensor is distorted by the distortion of the cylinder.

[Explanation of symbols] 1 Judgment unit 5 seat type pressure sensor 10 Pressure sensor element 50 spark plugs 51 metal shell 51f mounting seat SH cylinder head

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G01M 15/00 F02P 17/00 F (72) Inventor Koji Okazaki 14-18 Takatsuji-cho, Mizuho-ku, Aichi Prefecture No. Nihon Special Ceramics Co., Ltd. (72) Inventor Atsuji Ota No. 1 Toyota-cho, Toyota-shi, Aichi F-term inside Toyota Motor Corporation (reference) 2G087 AA13 BB14 CC12 EE24 FF08 3G019 CD01 DB07 GA01 GA02 GA15 KA28 KD17 3G084 BA00 BA16 DA20 DA28 EA01 EA07 EA11 EB06 EC02 FA21 FA39

Claims (4)

[Claims] 1. A cylinder internal pressure measurement value based on a cylinder internal pressure of an internal combustion engine to which the spark plug is mounted is obtained from an output of a pressure sensor provided on a mounting seat of the spark plug, and a crank angle is determined after the intake valve is closed. Determines the first pressure measurement time within a period until reaching the top dead center (hereinafter referred to as the top dead center period), and the in-cylinder pressure measurement value at the first pressure measurement time is P1, and the crank angle is A second pressure measurement time is set within a period from reaching the top dead center until the exhaust valve opens (hereinafter referred to as a top dead center period), and the in-cylinder pressure measurement value at the second pressure measurement time is set to P2. On the other hand, the correction measurement time is set within the period before the top dead center and before the ignition timing, and the in-cylinder pressure measurement value at the correction measurement time is set as the correction pressure value P0, and from the pressure measurement values P1 and P2, Before each The value obtained by subtracting the corrected pressure value P0 is P1 '.
And P2 ', the misfire determination method of the internal combustion engine is performed based on the ratio of P1' and P2 '. 2. The misfire determination of the internal combustion engine according to claim 1, wherein the first pressure measurement time and the second pressure measurement time are respectively set to times when the crank angles from the top dead center are equal to each other. Method. 3. The misfire determination method for an internal combustion engine according to claim 1, wherein the corrected measurement time is set to a time after the intake valve is closed. 4. A pressure sensor provided on a seat of the spark plug and a crank angle after the intake valve is closed to obtain a cylinder pressure measurement value based on a cylinder pressure of an internal combustion engine to which the spark plug is attached. A first pressure measurement time is set within a period until reaching the top dead center (hereinafter, referred to as a top dead center period), the in-cylinder pressure measurement value at the first pressure measurement time is set to P1, and the crank angle is set to the above A second pressure measurement time is set within a period from reaching the dead point to opening the exhaust valve (hereinafter, referred to as a top dead center period), and the in-cylinder pressure measurement value at the second pressure measurement time is set to P2, On the other hand, the correction measurement time is set within the period before the top dead center and before the ignition timing, and the in-cylinder pressure measurement value at the correction measurement time is set as the correction pressure value P0.
The value obtained by subtracting the corrected pressure value P0 from 1 and P2 is P
A misfire determination system for an internal combustion engine, comprising: a determination unit for determining the misfire of the internal combustion engine based on the ratio of P1 'and P2' as 1'and P2 '.