JP3427626B2 - Ignition timing control device for internal combustion engine - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ignition timing control device for an internal combustion engine, and more particularly to an ignition timing control device for stabilizing the engine.

[0002]

2. Description of the Related Art Conventionally, in an internal combustion engine, an air-fuel ratio sensor (specifically, an O ₂ sensor) is provided in the exhaust passage,
Based on the rich / lean signal, the air-fuel ratio feedback correction coefficient α is set by the well-known proportional-integral control, and the basic fuel injection amount Tp determined corresponding to the amount of air taken into the engine is set by the air-fuel ratio feedback correction coefficient α. By correcting and calculating the fuel injection amount Ti = Tp × α,
The air-fuel ratio is feedback-controlled to the theoretical air-fuel ratio.

However, due to the periodic rich / lean signal from the O ₂ sensor, the air-fuel ratio feedback correction coefficient α
Fluctuates cyclically, and even if the basic fuel injection amount Tp is constant, the fuel injection amount Ti becomes maximum when the air-fuel ratio feedback correction coefficient α is maximum, and the fuel injection amount when the air-fuel ratio feedback correction coefficient α is minimum. Ti is minimized. Therefore, fluctuations in the fuel injection amount Ti cause fluctuations in rotation, which are particularly noticeable during idle operation.

Therefore, in the ignition timing control apparatus disclosed in Japanese Patent Laid-Open No. 61-98970, as shown in FIG. 12, the deviation between the air-fuel ratio feedback correction coefficient α and the moving average value αave of the air-fuel ratio feedback correction coefficient is shown. (Α-αav
e) and, depending on this deviation, that is, if the deviation is positive, it is advanced by an amount according to its magnitude, and if the deviation is negative, it is advanced by an amount according to its magnitude. By correcting the ignition timing to the side, the idle rotation is stabilized.

[0005]

However, the conventional ignition timing control device for an internal combustion engine as described above has the following problems. Consider the actual air-fuel ratio of each cylinder of the internal combustion engine when the fuel injection amount changes. Figure
As shown in 13 (a), the air-fuel ratio feedback correction coefficient α
The step change of is set as Δα, and as a result,
As shown in, the fuel injected into a certain cylinder is ΔFuel =
When changing by Tp × Δα, a part of the fuel is taken by the increase of the wall flow, so that the fuel supplied to the cylinder changes along the broken line in the drawing for each intake frequency. The response characteristic of the intake fuel can be regarded as a first-order lag.

Therefore, as shown in FIG. 14 (b), the air-fuel ratio feedback correction coefficient is set as shown in FIG. 14 (b) with respect to changes in the air-fuel ratio of the O ₂ sensor section (sensor section air-fuel ratio) as shown in FIG. 14 (a). When α changes, the actual air-fuel ratio of each cylinder (in-cylinder air-fuel ratio) changes as shown in FIG. 14 (d). That is, the air-fuel ratio of each cylinder does not become the air-fuel ratio according to the change in α, but changes later than the change in α, and can be regarded as a first-order lag. Therefore, for example, when α changes from the decrease side to the increase side, the air-fuel ratio of each cylinder does not become rich immediately but remains lean.

In the ignition timing control device described in the above publication, since the ignition timing is corrected according to the air-fuel ratio feedback correction coefficient, the retard angle is retarded even though the actual air-fuel ratio of each cylinder is lean. However, there is a problem in that the rotation fluctuation rather increases due to the advance or the advance even though it is rich. In order to solve this point, in Japanese Patent Laid-Open No. 7-83150, a predetermined first order delay is added to the amount of change in the air-fuel ratio feedback correction coefficient calculated based on the signal from the O ₂ sensor provided in the exhaust passage. A method of calculating the air-fuel ratio in the combustion chamber of each cylinder with characteristics, and correcting the ignition timing according to the difference between the air-fuel ratio in the combustion chamber of each cylinder obtained by this and the theoretical air-fuel ratio has been proposed. There is.

However, it has been known that if there is a so-called valve deposit accumulated around the intake valve due to blowback to the intake pipe, the increase of the wall flow becomes extremely large. This is because the porous valve deposit absorbs the fuel when it is dry and releases the fuel when it is wet (including the fuel). As the degree of fouling (accumulation of valve deposit) increases, the first-order lag characteristic of the fuel response with respect to the change in the air-fuel ratio feedback correction coefficient α tends to become stronger.

In the ignition timing control device of Japanese Patent Laid-Open No. 7-83150 described above, since the first-order lag characteristic of the air-fuel ratio feedback correction coefficient is treated as constant, deterioration and dirt around the fuel injection valve progress and the fuel progresses. If the first-order lag characteristic of the response becomes strong, the error between the calculated air-fuel ratio and the actual air-fuel ratio of each cylinder may increase, and the ignition timing correction amount may be insufficient. In other words, there is a possibility that the amount of advancement may be insufficient even though a certain cylinder is lean, or that the amount of retardation may be insufficient even though it is rich, and rotation fluctuation may not be sufficiently suppressed. , There was a demand for improvement in this respect.

Further, particularly in an operating state of low rotation and high load, the ignition timing on the advance side is limited by knocking, so that the stability of the engine is apt to be impaired and the fuel injection amount is large. Wall flow increases and fuel response is significantly delayed. Conventionally, in such an operating state, the stability was improved by prohibiting the air-fuel ratio feedback control and fixing the air-fuel ratio to the rich side, but there was a problem that exhaust gas and fuel consumption deteriorated. .

The present invention has been made in view of such a conventional problem, and controls the ignition timing on the basis of the actual air-fuel ratio of each cylinder and the variation of the torque generated by each cylinder. By doing so, it is intended to ensure the stabilization of the institution.

[0012]

Therefore, in the invention according to claim 1, as shown in FIG. 1, the fuel is supplied based on a signal from an air-fuel ratio sensor which is provided in the exhaust passage and detects the air-fuel ratio of the exhaust gas. Air-fuel ratio feedback control means A for setting an air-fuel ratio feedback correction coefficient for correcting the injection amount and performing feedback control of the air-fuel ratio to the stoichiometric air-fuel ratio, and a first-order lag characteristic with respect to the amount of change in the air-fuel ratio feedback correction coefficient. In-cylinder air-fuel ratio estimating means B for calculating the air-fuel ratio in the combustion chamber of each cylinder, and the difference between the air-fuel ratio in the combustion chamber of each cylinder calculated by the in-cylinder air-fuel ratio estimating means and the theoretical air-fuel ratio. Ignition timing correction amount calculation means C for calculating the ignition timing correction amount according to the above, and the ignition timing correction amount calculated according to the ignition timing correction amount calculation means for correcting the ignition timing set according to the engine operating conditions. In an ignition timing control device for an internal combustion engine, which comprises an ignition timing correction means D, a generated torque detection means E for sequentially detecting a generated torque of each cylinder, and a torque for calculating a variation amount of the torque detected by the generated torque detection means. The magnitude of the first-order lag characteristic of the air-fuel ratio in the combustion chamber of each cylinder with respect to the amount of change in the air-fuel ratio feedback correction coefficient in the in-cylinder air-fuel ratio estimating means is calculated based on the variation calculation means F and the variation amount of the torque. By providing and compensating the primary delay characteristic correcting means G for correcting, the air-fuel ratio of 1
Precise ignition timing control that responds to changes in second-lag characteristics.

The above-described generated torque detecting means is configured to detect the generated torque of each cylinder based on the output of the in-cylinder pressure sensor arranged in each cylinder, as in the invention according to claim 2, for example. can do. The value obtained by integrating the output signal of the in-cylinder pressure sensor arranged in each cylinder over a predetermined angle has a strong correlation with the indicated mean effective pressure proportional to the generated torque of each cylinder, and therefore, indirectly. The generated torque is required.

Further, as in the invention according to claim 3, the torque generated in each cylinder may be detected based on the angular acceleration of the output shaft of the internal combustion engine, and the angular acceleration and the torque generated are in a proportional relationship, that is, The generated torque can be indirectly obtained by utilizing the characteristic that the angular acceleration increases as the generated torque increases. The angular acceleration of the output shaft of the internal combustion engine can be measured by using the internal combustion engine speed detection sensor.

When the torque fluctuation calculating means is configured to calculate the variance value of the torque detected by the generated torque detecting means as the torque fluctuation amount as in the invention according to claim 4, the torque fluctuation calculating means The amount of fluctuation can be accurately calculated. Further, the first-order lag characteristic correcting means increases the first-order lag characteristic of the air-fuel ratio when the fluctuation amount of the torque is larger than a predetermined stability threshold value, as in the invention according to claim 5. When the variation amount of the torque is equal to or less than the stability threshold value, the first-order delay characteristic of the air-fuel ratio is corrected to be reduced.

More specifically, as in the invention according to claim 6, with respect to the time constant T of the first-order lag characteristic represented by (1-e ^{−t / T} ) with respect to time t, When the variation amount of the generated torque is larger than the stability threshold value, a predetermined correction value is added, and when the variation amount of the generated torque is less than or equal to the stability threshold value, the predetermined correction value is subtracted to obtain an actual first-order delay. The ignition timing is corrected according to the characteristics.

[0017]

According to the first aspect of the present invention, the air-fuel ratio in the combustion chamber of each cylinder is calculated from the output of the air-fuel ratio sensor, the ignition timing is corrected based on the calculated air-fuel ratio, and the generated torque is detected and the fluctuation thereof is detected. By correcting the constant that calculates the air-fuel ratio in the combustion chamber of each cylinder according to the amount, it is possible to prevent hunting due to ignition timing correction, and reliably suppress rotational fluctuations during idling etc. effective.

As a result, the air-fuel ratio feedback control can be performed even under the operating conditions of low rotation and high load, and it is not necessary to fix the air-fuel ratio to the rich side to stabilize the engine. The effect that the performance and the fuel consumption can be improved is obtained. Further, according to the second aspect of the present invention, by estimating the generated torque for each cylinder based on the output signal of the in-cylinder pressure sensor, precise ignition timing correction control can be performed, hunting can be prevented, and rotation can be prevented. There is an effect that fluctuation can be surely suppressed.

According to the third aspect of the present invention, the torque for each cylinder is estimated from the angular acceleration of the crankshaft, so that inexpensive and precise ignition timing correction control can be performed, hunting can be prevented, and rotation can be prevented. There is an effect that fluctuation can be surely suppressed. Further, according to the invention of claim 4, there is an effect that a value reflecting the fluctuation of the generated torque can be easily obtained by calculating the variance value of the generated torque detected successively.

Further, according to the invention of claim 5, by comparing the variation amount of the generated torque with a predetermined stabilization threshold value,
There is an effect that the correction direction of the first-order lag characteristic can be determined easily and easily. According to the invention of claim 6, the time constant of the primary delay characteristic is adjusted based on the comparison between the generated torque and the predetermined value, so that the ignition timing is accurately adjusted according to the change of the primary delay characteristic. The effect is that correction control can be performed.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. 2 to 9 show an embodiment in which the ignition timing control device for an internal combustion engine according to the present invention is applied to a spark ignition type internal combustion engine with four cylinders and four cycles. Figure 2
FIG. 3 is a schematic configuration diagram showing an internal combustion engine and its peripheral devices in which fuel injection control and ignition timing control are performed. In order to detect the operating state of the internal combustion engine, a crank angle sensor 1 that measures the engine speed and the position of the crankshaft, and an air flow sensor 2 that is installed in the intake passage of the engine and that measures the intake air amount. , A water temperature sensor 3 for measuring the cooling water temperature of the internal combustion engine
And a throttle opening sensor 4 that is provided coaxially with the throttle valve to detect the throttle opening.
And an O ₂ sensor 5 for measuring the oxygen concentration in the exhaust gas, which constitute the operating state detecting means. Further, an in-cylinder pressure sensor 6 for independently detecting the in-cylinder pressure of each cylinder is provided in the ignition plug of each cylinder, and constitutes a generated torque detecting means. The outputs of these sensors are input to the internal combustion engine control unit 7.

The internal combustion engine control unit 7 determines the fuel injection amount Ti and the ignition timing A based on these operating state information.
It has a function of performing various calculations including DVn, and is composed of a CPU 8, a ROM 9, a RAM 10, and an I / O port 11. The CPU 8 fetches the required external data from the I / O port 11 according to the program written in the ROM 9, and performs the necessary processing while exchanging data with the RAM 10, and the I / O port according to the processing data. The signal is output from 11.
The ROM 9 stores a program for controlling the CPU 8, and the ROM 9 stores data used for calculation in the form of a map or a table. The signals from the sensor group are input to the I / O port 11 and the I / O port 11
From the fuel injection signal or ignition signal from the fuel injection valve 12,
Alternatively, an ignition coil unit independently arranged for each cylinder
It is output to 13.

The fuel injection valve 12 is arranged independently of each cylinder and supplies fuel corresponding to the input pulse width to the intake port of the internal combustion engine, and the ignition coil unit 13 sparks the spark plug 14 at a timing corresponding to the input signal. By doing so, the mixture in the cylinder is ignited. The exhaust gas after combustion is purified of harmful substances by the three-way catalyst 15. Next, the detailed operation of the present invention will be described based on the flowcharts shown in FIGS.

FIG. 3 is a flowchart of a fuel injection amount (Ti) calculation routine, which is executed at predetermined time intervals. First, step 1 (denoted as S1 in the figure. The same applies hereinafter).
Then, the intake air amount Qa detected based on the signal from the air flow sensor 2 is read, and the engine speed Ne detected based on the signal from the crank angle sensor 1 is read.
Read.

In step 2, from the intake air amount Qa and the engine speed Ne, the basic fuel injection amount Tp = K × Qa / Ne
(K is a constant) is calculated. In step 3, the basic fuel injection amount Tp is corrected by the current air-fuel ratio feedback correction coefficient α set by the air-fuel ratio feedback correction coefficient (α) calculation routine of FIG. The final fuel injection amount Ti is calculated.

Ti = Tp × α × COEF + Ts COEF is a correction coefficient including a water temperature correction coefficient, T
s is a voltage correction amount based on the battery voltage. When the fuel injection amount Ti is calculated, a drive pulse signal is output to the fuel injection valve with a width corresponding to this Ti at a predetermined timing synchronized with the engine rotation for each cylinder, and fuel injection is performed.

FIG. 4 is a flowchart of the air-fuel ratio feedback correction coefficient α calculation routine, which is executed at predetermined time intervals. This routine corresponds to air-fuel ratio feedback control means. In step 11, it is determined whether or not it is in the air-fuel ratio feedback control region (F / B region). The determination of the air-fuel ratio feedback control region is determined by the engine speed Ne, the basic fuel injection amount Tp, the cooling water temperature Tw, the O ₂ sensor output (active / inactive), and the like.

In the case of the air-fuel ratio feedback control region,
Go to step 12 to read the output voltage of the O ₂ sensor 5,
The rich / lean air-fuel ratio is determined by comparing with a slice level voltage corresponding to the theoretical air-fuel ratio. If the air-fuel ratio detected by the O ₂ sensor 5 is lean, step
The process proceeds to step 13 and it is determined whether the previous determination was rich.

In the case of rich last time, it is the time of reversal of rich → lean, so the routine proceeds to step 14, where it is now (just before reversal).
The air-fuel ratio feedback correction coefficient .alpha. Is stored as .alpha.old, and in step 15, the current (immediately before reversal) cylinder empty space calculated in step 46 of the ignition timing correction amount (.DELTA.ADVn) calculation routine described later in FIG. The fuel ratio equivalent value αAn is α
Remember as An-old. The subscript n represents the cylinder number,
Here, for n = 1 to 4, the current (immediately before reversal) in-cylinder air-fuel ratio equivalent value αAn is stored as αAn-old. still,
At this time, αAn-old may be uniformly set to a predetermined lean side air-fuel ratio equivalent value for n = 1 to 4.

Then, the routine proceeds to step 16, where for proportional control, a predetermined proportional amount P is added to the current air-fuel ratio feedback correction coefficient α, and the air-fuel ratio feedback correction coefficient α is added.
Is significantly increased. If it wasn't rich last time,
Since the lean state is continuing, the routine proceeds to step 17, where the current air-fuel ratio feedback correction coefficient α
Is added to a predetermined integral amount I (<< P) to update the air-fuel ratio feedback correction coefficient α to the minute amount increasing side.

When the air-fuel ratio detected by the O ₂ sensor 5 is rich, the routine proceeds to step 17, where it is judged if it was lean in the previous judgment. In the case of lean last time, it is during lean-> rich reversal, so the routine proceeds to step 19, where the current (immediately before reversal) air-fuel ratio feedback correction coefficient α is stored as αold, and in step 20, FIG.
The current (immediately before reversal) in-cylinder air-fuel ratio equivalent value αAn calculated in step 46 of the ignition timing correction amount (ΔADVn) calculation routine is stored as αAn-old. The subscript n represents a cylinder number. Here, for n = 1 to 4, the current (immediately before reversal) in-cylinder air-fuel ratio equivalent value αAn is stored as αAn-old. At this time, αAn-old is changed from n = 1 to
4 may be uniformly set to a predetermined rich side air-fuel ratio equivalent value.

Then, in step 21, for proportional control, a predetermined proportional amount P is subtracted from the current air-fuel ratio feedback correction coefficient α to obtain the air-fuel ratio feedback correction coefficient α.
Is significantly increased. If it wasn't lean last time,
Since the rich state is being continued, the routine proceeds to step 22, where the current air-fuel ratio feedback correction coefficient α
Is subtracted by a predetermined integral I (<< P) to update the air-fuel ratio feedback correction coefficient α to the minute amount reduction side.

If it is not in the air-fuel ratio feedback control range, this routine is terminated by the judgment in step 11,
At this time, the air-fuel ratio feedback correction coefficient α is clamped to the previous value. FIG. 5 is a flowchart of an ignition timing (ADVn) calculation routine, which is executed at predetermined time intervals.

In step 31, the engine speed Ne and the basic fuel injection amount Tp are read. In step 32, the map in which the basic ignition timing (basic ignition advance) MADV is determined according to the engine speed Ne and the basic fuel injection amount Tp is referred to, and MADV is searched from the actual Ne and Tp. In step 33, the cylinder to be ignited next (ignition cylinder; #n) is determined based on the cylinder determination signal of the crank angle sensor.

At step 34, the ignition timing correction flag FL is set.
The value of GAH (this is determined in the ignition timing correction amount calculation routine of FIG. 6 described later) is determined, and FLGAH =
If it is 1 (corrected), go to step 35. Step
At 35, the ignition timing correction amount ΔADVn corresponding to the ignition cylinder (#n) (this is calculated by the ignition timing correction amount calculation routine described later in FIG. 6) is added to the basic ignition timing MADV as in the following equation. Then, the final ignition timing (ignition timing) ADVn for the ignition cylinder (#n) is calculated.

ADVn = MADV + ΔADVn When FLGAH = 0 (no correction), step
At 36, the basic ignition timing ADV, the ignition timing MADV, and the final ignition timing ADVn for the ignition cylinder (#n) are unchanged.
And Although various other corrections are actually made,
It is omitted here.

In this way, when the final ignition timing ADVn for the ignition cylinder (#n) is determined, an ignition signal is output to the ignition coil unit 13 at that timing and the ignition cylinder (#n) is discharged. Ignition is performed by the spark plug 14. The steps 34 to 36 correspond to the ignition timing correction means. FIG. 6 is a flowchart for calculating the ignition timing correction amount (ΔADVn) as the ignition timing correction amount calculation means, which is executed at predetermined time intervals.

In step 41, it is judged whether or not it is in the air-fuel ratio feedback control region (F / B region). If it is in the air-fuel ratio feedback control region, the routine proceeds to step 42. Step
At 42, it is determined whether or not it is in the ignition timing correction region, and if it is in the ignition timing correction region, the routine proceeds to step 43 to calculate the ignition timing correction amount ΔADVn. The ignition timing correction may be performed in the entire air-fuel ratio feedback control region, but the correction region can be limited in order to shorten the calculation time of the internal combustion engine control unit 7 including other controls. In addition, a step of determining the correction area is provided.

At step 43, the ignition timing correction flag FL is set.
Set GAH to "1". In step 44, the cylinder (fuel injection cylinder; #n) for which fuel injection is to be performed next is determined based on the cylinder determination signal of the crank angle sensor 1. In step 45, αold stored in step 14 or step 19 of the α calculation routine of FIG. 4 described above, that is, the value of α (αold) immediately before the previous inversion of the air-fuel ratio feedback correction coefficient α is read, and the current value is read. The amount of change Δα = α−αold (FIG. 11) up to the air-fuel ratio feedback correction coefficient α is calculated.

In step 46, when αAn-old for the fuel injection cylinder (#n) stored in step 15 or step 20 of the α calculation routine of FIG. 4 described above, that is, when the air-fuel ratio feedback correction coefficient α is inverted last time The value (αAn-old) of the in-cylinder air-fuel ratio equivalent value αAn immediately before is read, and from this and the change amount Δα of the air-fuel ratio feedback correction coefficient α up to the present, according to the relationship of the first-order lag,
The current in-cylinder air-fuel ratio equivalent value αAn for the fuel injection cylinder (#n) is calculated.

ΑAn = αAn−old + Δα × (1-e ^{−t / T} ) T is a time constant, which is set by a delay constant correction routine of FIG. 9 described later. In general, t represents time, but here, particularly t represents the number of intakes of the fuel injection cylinder (#n) since the air-fuel ratio feedback correction coefficient α was inverted last time. As described above, since the in-cylinder air-fuel ratio changes with a first-order lag with respect to the target air-fuel ratio, the in-cylinder air-fuel ratio equivalent value αAn for the fuel injection cylinder (#n) is calculated by the above equation. The in-cylinder air-fuel ratio is represented by this. The steps 44 to 46 correspond to in-cylinder air-fuel ratio estimating means.

In step 47, as shown in FIG. 7, a map in which the ignition timing correction amount ΔADV is determined according to the in-cylinder air-fuel ratio equivalent value αAn is referred to, and the actually calculated fuel injection cylinder (#n) is calculated. The ignition timing correction amount ΔADVn of the cylinder is retrieved from the in-cylinder air-fuel ratio equivalent value αAn. The map used here has the characteristics shown in FIG. 7, and when the cylinder air-fuel ratio equivalent value αAn is positive, it is assumed that the cylinder air-fuel ratio becomes rich with respect to the theoretical air-fuel ratio, and the ignition timing is The ignition timing correction amount ΔADVn is set to a negative value having a magnitude corresponding to the absolute value of αAn so that the ignition timing is corrected to the retard side, and the cylinder air-fuel ratio equivalent value α
When An is negative, it is assumed that the in-cylinder air-fuel ratio has become leaner than the stoichiometric air-fuel ratio, and the ignition timing correction amount ΔADVn is set to a value corresponding to the absolute value of αAn so as to correct the ignition timing to the advance side. Sa positive value.

In this way, the ignition timing correction amount Δ for each cylinder
When ADVn is calculated, it is calculated as ADVn in FIG.
Used in the calculation routine, the ignition timing of the ignition cylinder #n is corrected. Therefore, when the air-fuel ratio feedback correction coefficient α changes as shown in FIG. 11A, the actual air-fuel ratio (cylinder A / F) of each cylinder is as shown in FIG.
Although it changes as shown in (b), this is estimated by calculation, and based on this, the ignition timing is corrected as shown in FIG. 11 (c).

If it is not in the air-fuel ratio feedback control region or in the ignition timing correction region, step
The routine proceeds to step 48, where the ignition timing correction flag FLGAH is reset to 0, and then this routine is ended. As a result, the ignition advance angle is corrected according to the air-fuel ratio in the cylinder to prevent rotation fluctuation. But 1
If the ignition timing is only corrected on the assumption that the secondary delay characteristic is constant, if the primary delay characteristic with respect to the air-fuel ratio feedback correction coefficient α changes due to deterioration or dirt near the intake valve, the actual air-fuel ratio is calculated. It is possible that the difference from the air-fuel ratio estimated by is increased and it becomes difficult to accurately correct the ignition timing.

Therefore, in the ignition timing control device according to the present invention, a routine for further detecting the torque generated in each cylinder,
A routine for correcting the time constant T used to calculate the in-cylinder air-fuel ratio equivalent value αAn is provided to correct the time constant T when calculating the in-cylinder air-fuel ratio equivalent value αAn based on the generated torque. The engine rotation is stabilized in response to changes in the next-delay characteristics.

FIG. 8 shows a flowchart of the generated torque detection routine. This routine is executed in synchronization with a unit signal (hereinafter referred to as a POS signal) output from the crank angle sensor 1 every 1 ° to 2 °, and corresponds to generated torque detecting means. First, in step 50, the level of a reference signal (hereinafter referred to as a REF signal) that becomes a Hi level is read from the crank angle sensor 1 for each crank angle of 180 °.

Next, at step 51, the REF signal level is judged. If the result is "Hi", it means the crankshaft reference position, so the routine proceeds to step 52, where the angle counter POSCNT and the cylinder pressure integration buffer ITGPC are used.
After clearing to zero, proceed to step 53. On the other hand, when the REF signal level is “Lo” in step 51,
Proceed to step 53 without going through step 52.

At step 53, the angle counter POSCNT
And the integration start angle TPCST are compared, and POSCNT ≧ TP
If it is CST, the process proceeds to step 54, and POSCNT
In the case of TPCST, it is determined that it is not in the in-cylinder pressure integration section, and the process ends. Next, at step 54, the angle counter POSCNT is compared with the integration end angle TPCED. If the result is POSCNT ≠ TPCED, the process proceeds to step 55.

At step 55, the angle counter POSCN is set.
T is compared with the integration end angle TPCED, and POSCNT <
In the case of TPCED, since it is the integration section of the in-cylinder pressure, the routine proceeds to the next step 56, where the in-cylinder pressure Pc is read from the output signal of the in-cylinder pressure sensor 6. Then, the routine proceeds to step 58, where the in-cylinder pressure Pc is stored in the in-cylinder pressure integration buffer ITGPC.
Is added and the process ends.

At step 54, POSCNT = TP
In the case of CED, since it is at the end angle of the in-cylinder pressure integration section, the routine proceeds to step 57, where the value of the in-cylinder pressure integration buffer ITGPC is substituted into the simplified cylinder-specific Pin (integrated value of each in-cylinder pressure). Then, the process ends. If POSCNT> TPCED in step 55, the integration section of the in-cylinder pressure has already ended, so the process ends.

Since the simplified cylinder-specific Pin at the predetermined angle, particularly the combustion stroke, obtained by the routine of FIG. 8 is proportional to the indicated mean effective pressure Pi, assuming that the pump loss in each stroke is equal, it is large. It can be regarded as equivalent to the detection of the indicated mean effective pressure Pi, that is, the generated torque without the need for calculation processing. The simple cylinder Pins obtained in this routine are sequentially stored in the buffer memory in the order of sampling.

Next, FIG. 9 shows a flowchart of a time constant correction routine used to calculate the in-cylinder air-fuel ratio equivalent value αAn. This routine is executed for each predetermined crank reference position. First, at step 60, the simplified cylinder-specific Pin for each cylinder (#n) obtained in the routine of FIG. 8 and sequentially stored in the buffer memory in the order of sampling is read.

Next, at step 61, all cylinder average value AAVP
Calculate i. The average value is calculated by the following formula based on the latest value of P in using the weighted average method. AAVPi = KAVE × AAVPi ₀ + (1-KAV
E) × Pin where KAVE is an average weighting coefficient and 0 <KAVE ≦ 1.
AAVPi ₀ is a constant in the range of
This is the previous value of VPi.

This all-cylinder average value AAVPi is the sum of the simplified cylinder-specific Pins for each cylinder stored in the buffer memory, and a correction calculation cycle REFSL described later, that is,
A value obtained by dividing by the sampling number may be used. In the next step 62, it is determined whether or not the # 2 cylinder is determined from the current REF signal, that is, whether or not the simplified cylinder Pins for all the cylinders are obtained. Since the Pin detection routine of FIG. 8 obtains the detection position based on the REF signal,
In a 4-cylinder engine in which the combustion stroke is performed in the order of 1 cylinder → # 3 cylinder → # 4 cylinder → # 2 cylinder, if the # 1 cylinder is set as the Pi detection start cylinder, the # 1 cylinder enters the combustion stroke again and REF If the signal is the reference signal corresponding to the combustion stroke of the # 1 cylinder, it means that the previous Pi detection (up to the # 2 cylinder) has been completed, and at this point, the simplified cylinder Pins for all the cylinders are obtained. It will be.

As a result of the determination in step 62, if the simple cylinder Pins for all the cylinders have been obtained, the process proceeds to step 63, and if not, the process ends. In step 63, the sampling cycle counter REFCNT is compared with a predetermined correction calculation cycle REFSL, and REFCNT ≧
If it is REFSL, go to step 64, REFCNT
<If REFSL, proceed to step 65. For this reason,
A predetermined number of simple average effective pressures P in of the # 1 to # 4 cylinders are sampled, and the subsequent correction processing is performed at the time when the all-cylinder average value AAVPi is obtained.

In step 64, the sampling cycle counter REFCNT is cleared to zero for initialization, and in step 66
Go to. On the other hand, in step 65, the sampling cycle counter REFCNT is incremented and the processing is ended.
In step 66, the variance value SPi is calculated from the previously calculated average value AAVPi of all cylinders and the simple cylinder-specific Pin. The variance value SPi is calculated by the following formula.

SPi = {(Pi1-AAVPi) ² +
(Pi2-AAVPi) ² + ... + (Pim-AA
VPi) ² } / (m-1) The sampling sequence of the simple average effective pressure required for this calculation is stored and stored in the buffer memory in the order of sampling in step 61 as described above.
When detection is performed REFSL times for each cylinder up to # 4, there are a total of m = 4 × REFSL sampling data.

Steps 60 to 66 correspond to torque fluctuation calculating means. Next, at step 67, the value of the ignition timing correction flag FLGAH determined by the routine of FIG. 6 is determined. When FLGAH is "1", the routine proceeds to the next step 68 to correct the ignition timing, and when it is "0", it is determined that it is not in the correction region, and the processing of this routine is ended.

In step 68, the dispersion value SPi and the stability threshold value KADVSL are used to determine the direction of ignition timing correction.
Compare with. If the result is SPi> KADVSL, the process proceeds to the next step 69, and if not, the process proceeds to step 70. In step 69, since the dispersion value SPi exceeds the stability threshold value KADVSL, a predetermined time constant correction gain KDYGIN is added to the time constant T in order to increase the time constant for calculating the cylinder air-fuel ratio equivalent value.

On the other hand, in step 70, since the dispersion value SPi does not exceed the stability threshold value KADVSL, the time constant T is subtracted from the time constant correction gain KDYGIN in order to reduce the time constant for calculating the cylinder air-fuel ratio equivalent value. To do. The steps 67 to 70 correspond to the first-order delay characteristic correction means. After performing the above processing, this routine ends.

In this way, the dispersion value of a value proportional to the generated torque is calculated, and this dispersion value is compared with a predetermined value. If this dispersion value exceeds the predetermined value, the wall flow rate is increasing. Correct the time constant for calculating the in-cylinder air-fuel ratio equivalent value to a smaller value. Therefore, for example, even if the first-order delay becomes large as shown by the broken line in FIG. 11 (b), the ignition timing is set to a correction value based on the time constant at which the stability becomes the best as shown by the broken line in FIG. 11 (c). Since it is corrected, the stability of the engine can always be kept good.

The generated torque detection routine of FIG. 8 can be replaced by the routine shown in FIG. First,
The principle of detecting the generated torque in the routine of FIG. 10 will be described. The angular acceleration Δω for each cylinder measured in the combustion stroke is
As shown in the following equation, it is proportional to the torque T generated in the combustion cylinder. T = Pi−Tf∝Δω = dω / dt where Pi is the indicated mean effective pressure, Tf is the friction torque,
dω is the angular velocity change amount, and dt is the sample time.

Therefore, if the variance value is calculated based on this angular acceleration, the variance value of the generated torque is indirectly, that is,
The engine stability can be obtained. The flowchart shown in FIG. 10 is called and executed for each unit crank angle (for example, for each POS signal). First, in step 80, in order to determine whether or not it is in the first angular velocity measurement section, the elapsed angle counter POSN from the crank reference position (REF signal) based on the output of the crank angle sensor 1 and the first angular velocity measurement section Is compared with the starting angle OMGS1. At this time, if the elapsed angle counter POSN is smaller than OMGS1, the current crank angle has not reached the first measurement section, so the routine proceeds to step 90, where the elapsed angle counter PO
Increment SN.

When POSN = OMGS1 in step 80, since it is at the start position of the first angular velocity measurement section, the elapsed time timer TMPOS is cleared and set to 0 in step 81, and in step 90 the elapsed angle counter POSN Is incremented. Elapsed time timer TMP
The OS is a free-running timer built into the internal combustion engine control unit, and continues to count up in synchronization with the internal clock unless the contents are cleared, so that the required time can be accurately measured regardless of the processing load of the internal combustion engine control program. It is possible.

In step 80, POSN> OMGS1
In the case of, in order to determine whether or not the first angular velocity measurement section has ended, the routine proceeds to step 82, where the elapsed angle counter P
The OSN is compared with the end angle OMGE1 of the first angular velocity measurement section. When POSN <OMGE1, the first angular velocity measurement section has not ended, so the routine proceeds to step 90, where POSN is incremented.

When POSN = OMGE1, since it is the end position of the first angular velocity measurement section, the content of the measurement section elapsed time timer TMPOS is loaded into the first elapsed time measurement value T1i in step 83, and the routine proceeds to step 90, where POSN is incremented. To do. If POSN <OMGE1 in step 82, the process proceeds to the next step 84 to determine whether it is in the second angular velocity measurement section, and the elapsed angle counter POSN and the start angle OMGS2 of the second angular velocity measurement section are set. To compare.

If POSN <OMGS2 in step 84, it means that the second angular velocity measurement section has not been reached, so the routine proceeds to step 90, where the elapsed angle counter POSN is incremented. When POSN = OMGS2,
Since it is at the start position of the second angular velocity measurement section, the elapsed time timer TMPOS is cleared to 0 in order to start timing.
, And at step 90, the POSN is incremented.

Then, in the case of POSN> OMGS2, in order to determine whether or not the second angular velocity measurement section is finished, the routine proceeds to step 86, where the elapsed angle counter POSN and the second angular velocity measurement section end angle OMGE2 are set. To compare.
When POSN <OMGE2, the second angular velocity measurement section has not ended, so the routine proceeds to step 90, where POSN
Is incremented.

When POSN = OMGE2,
Since it is the end position of the second angular velocity measurement section, the content of the measurement section elapsed time timer TMPOS is loaded in the second elapsed time measurement value T2i in step 87. Then the next step 88
Then, the difference between the second elapsed time measurement value T2i and the first elapsed time measurement value T1i obtained in step 83 is obtained, and step 90
Go to and increment the POSN. This T2i-
T1i is a variable corresponding to angular acceleration, and can be treated as a simple cylinder-specific Pin from the above-mentioned relationship.

Then, when POSN> OMGE2, the routine proceeds to the next step 89, where the REF signal level is judged. If the REF signal level is "Lo", the elapsed time counter POSN is incremented in step 90. On the other hand, when the REF signal is at the "Hi" level, since it is at the start position of the next cycle, the elapsed angle counter POSN is cleared to zero in step 91, and this routine is ended.

With the above processing, the difference in elapsed time between the two angular velocity measurement sections can be measured during one stroke. This corresponds to the reciprocal of the angular acceleration and is proportional to the simplified cylinder-specific Pin. Therefore, the routine of FIG. 9 can be executed based on this value to obtain the correction value of the time constant. Although it is possible to directly calculate the angular acceleration and perform the same processing, the processing by the routine of FIG. 10 has a light calculation load and the obtained effect remains unchanged.

In the above example, all four cylinder internal combustion engines have been described, but similar control is possible with N cylinder internal combustion engines.

[Brief description of drawings]

FIG. 1 is a functional block diagram showing the configuration of the present invention.

FIG. 2 is a system diagram showing an embodiment of the present invention.

FIG. 3 is a flowchart of a fuel injection amount calculation routine.

FIG. 4 is a flowchart of an air-fuel ratio feedback correction coefficient calculation routine.

FIG. 5 is a flowchart of an ignition timing calculation routine.

FIG. 6 is a flowchart of an ignition timing correction amount calculation routine.

FIG. 7 is a diagram showing a map for calculating an ignition timing correction amount.

FIG. 8 is a flowchart of a generated torque detection routine.

FIG. 9 is a flowchart of an ignition timing correction constant correction calculation routine.

FIG. 10 is a flowchart showing another embodiment of the generated torque detection routine.

FIG. 11 is a diagram showing a correction characteristic of ignition timing.

FIG. 12 is a diagram showing a conventional ignition timing correction characteristic.

FIG. 13 is a diagram showing a characteristic of cylinder intake fuel with respect to a step change of an air-fuel ratio feedback correction coefficient.

FIG. 14 is a diagram showing how the in-cylinder air-fuel ratio changes.

[Explanation of symbols]

1 Crank Angle Sensor 2 Air Flow Sensor 3 Water Temperature Sensor 4 Throttle Sensor 5 O ₂ Sensor 6 Cylinder Pressure Sensor 7 Internal Combustion Engine Control Unit 8 CPU 9 ROM 10 RAM 11 I / O Port 12 Fuel Injection Valve 13 Ignition Coil Unit 14 Spark Plug

Continuation of the front page (51) Int.Cl. ⁷ identification code FI F02D 45/00 362 F02D 45/00 364B 364 368S 368 F02P 5/15 K (56) Reference JP-A-3-124940 (JP, A) ( 58) Fields surveyed (Int.Cl. ⁷ , DB name) F02P 5/15 F02D 41/04 305 F02D 41/14 310 F02D 43/00 301 F22D 45/00 362 F02D 45/00 364 F02D 45/00 368

Claims (6)

(57) [Claims] 1. An air-fuel ratio feedback correction coefficient for correcting a fuel injection amount based on a signal from an air-fuel ratio sensor provided in an exhaust passage for detecting an air-fuel ratio of exhaust gas,
Air-fuel ratio feedback control means for feedback-controlling the air-fuel ratio to the stoichiometric air-fuel ratio, and 1 for the amount of change in the air-fuel ratio feedback correction coefficient.
In-cylinder air-fuel ratio estimating means for calculating the air-fuel ratio in the combustion chamber of each cylinder with the next-delay characteristic, and the air-fuel ratio and the theoretical air-fuel ratio in the combustion chamber of each cylinder calculated by the in-cylinder air-fuel ratio estimating means Ignition timing correction amount calculation means for calculating the ignition timing correction amount according to the difference between the ignition timing correction amount and the ignition timing correction amount calculated by the ignition timing correction amount calculation means for correcting the ignition timing set according to the engine operating condition. In an ignition timing control device for an internal combustion engine, comprising: ignition timing correction means; generated torque detection means for sequentially detecting generated torque of each cylinder; and torque fluctuation for calculating a fluctuation amount of torque detected by the generated torque detection means. A first delay of the air-fuel ratio in the combustion chamber of each cylinder with respect to the amount of change of the air-fuel ratio feedback correction coefficient in the in-cylinder air-fuel ratio estimating device based on the amount of change in the torque. Ignition timing control apparatus for an internal combustion engine, wherein a primary delay characteristic correcting means for correcting the magnitude of the characteristic, that the provided. 2. The ignition of an internal combustion engine according to claim 1, wherein the generated torque detecting means detects the generated torque of each cylinder based on the output of an in-cylinder pressure sensor arranged in each cylinder. Timing control device. 3. The ignition timing control device for an internal combustion engine according to claim 1, wherein said generated torque detecting means detects the generated torque of each cylinder based on the angular acceleration of the output shaft of the internal combustion engine. 4. The torque fluctuation calculation means calculates a variance value of the torque detected by the generated torque detection means as a fluctuation amount of the torque. An ignition timing control device for an internal combustion engine according to any one of claims. 5. The first-order lag characteristic correcting means corrects the first-order lag characteristic of the air-fuel ratio in a direction of increasing it when the amount of fluctuation of the torque is larger than a predetermined stability threshold value. The internal combustion engine according to any one of claims 1 to 4, wherein when the variation amount is equal to or less than the stability threshold value, the first-order lag characteristic of the air-fuel ratio is corrected to be reduced. Ignition timing control device. 6. The variation amount of the generated torque with respect to the time constant T of the first-order lag characteristic represented by (1-e ^{−t / T} ) with respect to time t, Is larger than the stability threshold, a predetermined correction value is added, and when the variation amount of the generated torque is equal to or smaller than the stability threshold, the predetermined correction value is subtracted.
An ignition timing control device for an internal combustion engine according to claim 5.