JP2870286B2 - Air-fuel ratio control device for lean burn 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 air-fuel ratio control device for a lean burn engine.

[0002]

2. Description of the Related Art A lean burn gasoline engine (lean burn engine) has two excellent characteristics of low fuel consumption and low pollution (Japanese Patent Application Laid-Open No. 61-232364, or
Automotive Engineering Society, Academic Lecture Preprints 911, 1991-5
See page 137).

When a conventional engine is operated at an air-fuel ratio leaner than the stoichiometric air-fuel ratio, combustion becomes unstable at an air-fuel ratio of around 19, and torque fluctuation increases, making it difficult to perform smooth operation. Become.

[0004] Although the fuel efficiency can be improved to some extent by operating at a lean air-fuel ratio, NOx cannot be sufficiently reduced due to unstable combustion. This is because the heat capacity of the combustion gas increases as the air-fuel ratio becomes leaner, so that the leaner the air-fuel ratio, the lower the combustion temperature and the smaller the NOx emission, but the leaner the air-fuel ratio, the more stable the combustion. Because it becomes worse
This is because the air-fuel ratio cannot be made lean enough to sufficiently reduce NOx.

[0005] In this case, if the lean combustion limit is improved by improving the combustion such as giving swirl to the intake air flowing into the combustion chamber, NOx can be further reduced, thereby satisfying the NOx regulation value. Significant improvement in fuel efficiency can be achieved.

In the lean burn engine, a swirl control valve and a swirl port are provided to improve combustion in a lean air-fuel ratio range, so that swirl control is performed, and a fuel injection amount given from an injector is independently controlled for each cylinder. In order to accurately maintain the air-fuel ratio within a range where both NOx reduction and torque fluctuation suppression are achieved, the air-fuel ratio feedback control is performed based on the lean mixture sensor with the air-fuel ratio on the lean side as a target value.

[0007] Lean combustion performance such as NOx emissions and torque fluctuations is affected by variations in the engine and air-fuel ratio control system such as a lean mixture sensor during mass production, and environmental conditions such as humidity. For this reason, the target air-fuel ratio in the air-fuel ratio feedback control using the lean mixture sensor must be set with a margin on the rich side of the torque fluctuation allowable limit. Therefore, it is effective to reduce this margin to reduce NOx.

[0008] Here, there is a correlation between the rotation fluctuation of the engine and the torque fluctuation when the air-fuel ratio changes, and both the rotation fluctuation and the torque fluctuation increase as the air-fuel ratio becomes leaner.

In view of this, an allowable limit of the torque fluctuation is detected on the vehicle from the rotation fluctuation of the engine, and the target value of the air-fuel ratio provided on the lean side is corrected so that the maximum value of the rotation fluctuation becomes equal to or less than the allowable value. Thus, the engine can be operated near the permissible torque fluctuation limit. For example, FIG. 13 shows the relationship between the rotational fluctuation rate and the air-fuel ratio when the load is constant (5th speed, 60 km / h, equivalent to road load), and the rotational fluctuation rate increases as the air-fuel ratio becomes leaner. When the actual rotation fluctuation rate exceeds the determination criterion indicated by the broken line, it is determined that combustion is unstable, and the air-fuel ratio target value is corrected to the rich side so that the air-fuel mixture becomes rich.

[0010]

However, since the rotational fluctuation is also caused by the fluctuation of the load, comparing the rotational fluctuation rate with a predetermined judgment criterion as described above, the load fluctuation becomes an error and the combustion is performed. The accuracy of determining whether or not it is unstable is reduced. For example, when there is no load fluctuation and the rotation fluctuation rate does not exceed the criterion, the rotation fluctuation generated due to the load fluctuation caused by the throttle operation is added, whereby the rotation fluctuation rate exceeds the criterion. Also at this time, it is determined that combustion is unstable, and the air-fuel ratio is enriched.

[0011] However, in this case, unless the throttle is operated, the determination standard is not exceeded.
There is an error in determining that combustion is unstable. If the air-fuel ratio is enriched even though the combustion is not unstable due to the erroneous determination, the fuel consumption will be increased wastefully, resulting in poor fuel economy and increased NOx emission.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to eliminate the influence of a load fluctuation when judging combustion instability by correcting a criterion or a rotational fluctuation rate of an engine in accordance with an engine load fluctuation rate. .

[0013]

According to the first aspect of the present invention, as shown in FIG. 1, when a condition that enables operation at an air-fuel ratio leaner than the stoichiometric air-fuel ratio is satisfied, the air-fuel ratio on the lean side is set as a target value. Means 32 for feedback-controlling the amount of fuel supplied to the engine so that the actual air-fuel ratio matches the value, and means 33 for calculating the load variation rate from the detected value of the engine load.
A means 34 for setting a criterion for the rotation fluctuation rate based on the load fluctuation rate, a means 35 for calculating the rotation fluctuation rate from the detected value of the engine speed, and comparing the rotation fluctuation rate with the criterion. Means 36 and means 37 for correcting the target value to the side where the air-fuel mixture becomes richer when the rotation fluctuation rate is larger than the determination reference based on the comparison result are provided.

According to the second invention, as shown in FIG. 2, when the condition that the air-fuel ratio on the lean side of the stoichiometric air-fuel ratio can be satisfied is satisfied, the air-fuel ratio on the lean side is set as the target value and the actual air-fuel ratio Means 32 for feedback-controlling the amount of fuel supplied to the engine so that the values of the engine speed coincide with each other, means 33 for calculating the load fluctuation rate from the detected value of the engine load, and means 35 for calculating the rotation fluctuation rate from the detected value of the engine speed And the ratio between the rotation fluctuation rate and the load fluctuation rate plus a positive constant value (for example, rotation fluctuation rate / (1 + load fluctuation rate))
41, means 42 for comparing the ratio with a predetermined criterion, and means 43 for correcting the target value to a side where the mixture becomes richer when the ratio is greater than the criterion based on the comparison result. Provided.

According to a third aspect of the present invention, in the first or second aspect, means is provided for correcting the criterion so that the criterion becomes larger as the gear position becomes lower according to the gear position.

According to a fourth aspect of the present invention, in the first or second aspect, the comparison with the criterion is performed by using a variation range of a load or a rotation speed.
Means is provided only when the engine is in a steady state in which at least one of the fluctuation widths is equal to or less than a predetermined value .

[0017]

The reason why the engine rotation fluctuation rate is compared with a predetermined judgment criterion is to judge the combustion instability caused by the variation of the air-fuel ratio sensor or the like when performing feedback control with the lean air-fuel ratio as a target value. To do that.

However, the engine rotation fluctuation rate is greatly affected by the load fluctuation, and the rotation fluctuation rate increases as the load fluctuation rate increases even at the same air-fuel ratio.

In this case, in the first aspect, the criterion for determining the rotational fluctuation rate based on the load fluctuation rate is set to be larger by, for example, the rotational fluctuation caused by the load fluctuation. That is, when the rotation fluctuation rate increases due to the load fluctuation, the determination criterion is increased by the same amount as the rotation fluctuation rate increases.

Thus, even when there is a load fluctuation, the result of comparison between the rotation fluctuation rate and the determination criterion is the same as when there is no load fluctuation, so that the determination of unstable combustion is stabilized.

In the second invention, the ratio between the rotation fluctuation rate and the load fluctuation rate plus a positive constant value is compared with a criterion.

In this case, since the load fluctuation rate and the rotation fluctuation rate to which the load fluctuation is added are in a proportional relationship, if the rotation fluctuation rate of the numerator increases as the load fluctuation rate increases, the value of the denominator becomes the same at the same rate. Also increases. In other words, the denominator and the numerator increase at the same rate according to the load change,
The above ratio is a constant value.

This also makes the comparison result between the above ratio and the judgment criterion the same irrespective of the presence or absence of a load change, thereby stabilizing the judgment of unstable combustion.

The effect of the load fluctuation on the rotation fluctuation rate differs depending on the gear position. Even at the same air-fuel ratio and the same load fluctuation, the rotation fluctuation rate increases at lower speeds. In this case, in the third invention, if the correction is made so that the determination criterion becomes larger as the speed becomes lower, the comparison result of the rotation fluctuation rate and the determination criterion in the first invention becomes irrespective of the gear position. The comparison result of the ratio and the determination criterion in the second invention is the same.

As a result, the accuracy of the determination of combustion instability is improved even in a low-speed stage where the influence of the load fluctuation is large.

In the fourth invention, the comparison with the criterion is based on the load.
At least one of the fluctuation width and the rotation speed fluctuation width is a predetermined value
When limited to the following steady state of the engine, the calculated values of the rotational fluctuation rate and the load fluctuation rate are more stable than in the first and second inventions.

[0027]

[Embodiment] In FIG. 3, in the lean burn engine,
While maintaining the air-fuel ratio of the air-fuel mixture flowing into the combustion chamber at a value leaner than the stoichiometric air-fuel ratio, the fuel efficiency is improved, while the NO that increases due to combustion instability in the lean air-fuel ratio region is increased.
In order to suppress x, swirl is given to the intake air flowing into the combustion chamber when combustion is unstable such as at a light load. A swirl control valve 13 partially notched (not shown) is provided near the intake port 12 to detect a light load state from an intake air amount signal (when an idle state is detected from a signal from the throttle sensor 6). Also, by swirling the intake air with the swirl control valve 13 in the fully closed position, the flow velocity of the intake air is increased, and swirl is generated in the combustion chamber.

The swirl control valve 13 is driven by a diaphragm actuator 14. When the swirl control solenoid 15 is turned on by a signal from the control unit 2, a negative pressure (operating pressure) from a vacuum tank 16 is guided to the actuator 14. ,
When the valve 13 is closed and the solenoid 15 is turned off, the valve 13 returns to the fully open position.

On the other hand, in the lean burn engine, there is a limit to the output generated at the target air-fuel ratio on the lean side. When a higher output is required, the actual air-fuel ratio needs to be on the rich side, which increases the NOx emission. In order to cope with this, when the operating range requires a high output, the target air-fuel ratio is switched to the stoichiometric air-fuel ratio, and the exhaust pipe 1
NOx is purified by the three-way catalyst 19 provided in the fuel cell 8.

Control unit 2 composed of microcomputer
In order to switch the target air-fuel ratio, a target excess fuel ratio TDML defined by TDML = 1 / (target air-fuel ratio / theoretical air-fuel ratio) is introduced, and a fuel injection pulse width Ti to the injector 3 provided in the intake port 12 is introduced. Where: Tp: basic injection pulse width determined from the intake air amount and the engine speed; COEF: 1 and the sum of various correction coefficients such as water temperature increase α: air-fuel ratio feedback correction coefficient Ts: Calculated by the invalid pulse width according to the battery voltage.

When switching from the target air-fuel ratio on the lean side to the stoichiometric air-fuel ratio, if the stoichiometric air-fuel ratio is set in the target air-fuel ratio in the equation, TDML = 1. The equation at this time is Ti = Tp · COEF · α + Ts, which is a known equation given to a conventional three-way catalyst system.

Conversely, when switching to the target air-fuel ratio on the lean side, a target value TDML predetermined for the target air-fuel ratio is set.
By inserting 0 (for example, 22), the TDML becomes 14.7 / 22 (that is, a value smaller than 1), and the TDML reduces the fuel supply amount at a stroke as compared with the case where the stoichiometric air-fuel ratio is targeted. By introducing TDML and changing the fuel amount stepwise at the time of switching, the convergence at the time of switching to the lean target air-fuel ratio is improved. Thereafter, the actual air-fuel ratio is detected by the wide area air-fuel ratio sensor 7, and the air-fuel ratio feedback control is performed so that the actual air-fuel ratio matches the lean-side target air-fuel ratio.

As described above, the control unit 2 for performing the switching control of the target air-fuel ratio and the feedback control of the air-fuel ratio.
Are input from various sensors for detecting the operating conditions of the engine required for control. Reference numeral 4 denotes an air flow meter for detecting the amount of air sucked from the air cleaner, 6 denotes a throttle sensor for detecting the opening of the throttle valve 5, and 7 outputs a signal for each unit crank angle and a signal for each crank angle reference position. A crank angle sensor, 8 is a water temperature sensor, and 9 is a wide-range air-fuel ratio sensor capable of widely detecting an actual air-fuel ratio from a stoichiometric air-fuel ratio to a lean air-fuel ratio.

When the feedback control is performed to the lean target air-fuel ratio, the combustion becomes unstable due to the variation of the air-fuel ratio sensor 9 and the like, thereby causing a torque fluctuation. Control (lean limit control) for correcting the air-fuel ratio target value so that the engine is operated near the permissible limit of fluctuation is performed.

In this case, if the actual rotation fluctuation rate related to the torque fluctuation is compared with a predetermined judgment criterion, and if this is larger than the judgment criterion, it is judged that the combustion is unstable. When the accompanying rotation fluctuation is added, it may be erroneously determined that the combustion is unstable.

In order to avoid this, the control unit 2 corrects the rotational fluctuation rate by (1 + load fluctuation rate) and compares the corrected value with a predetermined judgment criterion, so that an error due to load fluctuation is included. Not to be.

More specifically, FIGS. 4 and 5 are flowcharts for performing lean limit control, which are executed every two revolutions of the engine. Initially, the following conditions: basic injection pulse width (engine load equivalent amount) Tp is within a predetermined range; engine speed Ne is within a predetermined range; vehicle speed VSP detected by vehicle speed sensor 10 is within a predetermined range. It is determined whether or not all of the conditions (ie, GO conditions in the figure) are satisfied (step 1). If all of these conditions are satisfied, it is determined that the condition for performing the lean limit control has been satisfied. Proceeding to 2 and later, Tp and Ne are sampled by a fixed number.

However, the minimum value and the maximum value of the sample value during the sampling period are obtained, and if the difference between the minimum value and the maximum value does not fall below a predetermined value, it is determined that the time is not a steady state,
Does not perform lean limit control. For example, every time Tp is sampled, it is compared with the values stored in Tpmin and Tpmax of the memory, and if Tp ≦ Tpmin, Tp is set to Tp.
By putting Tp in Tpmax and Tp ≧ Tpmax, the minimum value during the sampling period is given to Tpmin.
The maximum value during the sampling period is entered in Tpmax (steps 2 to 5). Similarly, for Ne, the minimum value is stored in Nemin of the memory and the maximum value is stored in Nemax of the memory (steps 6 to 9). Using these maximum and minimum values, Tpmax−Tpmin is within a predetermined value Tpw and Nemax
When −Nemin is also within the predetermined value New, it is determined that the time is a steady state (steps 10 and 11), and the process proceeds to step 12 and thereafter. Here, Tpmax−Tpmin ≦ Tpw and Nem
The case where both ax−Nemin ≦ New is satisfied is determined to be the steady state. However, when only one of the conditions is satisfied, the steady state may be determined.

In a steady state throughout the sampling period, N memories Tp (1),..., Tp (n),.
p (N) and N memories Ne (1),..., Ne (n),
.. Tp and Ne enter Ne (N) in the sampling order (steps 12 to 14). N is the number of samples (about 50
).

Once all N sample values have been obtained,
Average value for Tp and Ne

(Equation 1) And variance σ _TP , σ _Ne

(Equation 2) Is calculated (steps 16 to 19).

When these calculations are completed,

(Equation 3) Rotation fluctuation rate defined by

(Equation 4) Is calculated from the load fluctuation rate defined in (1), and this is compared with the judgment criterion Limit (step 20), whereby the rotation fluctuation rate / (1 + load fluctuation rate) ≧ Limit If so, it is determined that combustion is unstable.

To judge combustion instability, the rotational fluctuation rate /
The reason for adopting the value of (1 + load variation rate) is as follows.
Now, when the air-fuel ratio is constant, as shown in FIG. 7 (when the air-fuel ratio is 22), the variable on the horizontal axis is (1 + load fluctuation rate), and the vertical axis is the rotation fluctuation rate. In proportion to this, the rotation fluctuation rate on the vertical axis also increases.

In the figure, when the variable on the horizontal axis is 1, there is no load fluctuation. On the other hand, when the variable on the horizontal axis becomes x having a value slightly deviated from 1 (when there is a small load fluctuation), the rotational fluctuation rate is increased by ba in the figure due to the influence of the load fluctuation, When the variable of the axis becomes y having a value larger than x (when there is a large load change), the load change increases the rotational fluctuation rate by ca. That is, in the state where there is no load fluctuation, when the rotation fluctuation rate exceeds the determination criterion Limit, all of the rotation fluctuation is due to combustion instability. However, in the state where there is a load fluctuation, the rotation fluctuation due to the load fluctuation is added as an error. Just because the rate exceeds the criterion Limit is not necessarily due to combustion instability.

In this case, since the load fluctuation rate and the rotation fluctuation rate to which the load fluctuation is added are in a proportional relationship, the rotation fluctuation rate /
When the value of the load change rate is adopted, the denominator,
This value is the same regardless of the presence or absence of the load fluctuation, since the numerator also increases at the same rate. That is, the effect of the load fluctuation is eliminated.

However, in actual control, the rotational fluctuation rate /
The value of (1 + load fluctuation rate) is adopted.

This value, which shows the slope (constant) of the straight line in FIG. 7, remains the same regardless of the load fluctuation. In this case, the reason why the value added to the load change rate of the denominator is set to 1 is that when the load change rate is 0, the denominator is set to 1 so that the value is easy to handle and the detection accuracy is improved. Although a positive number other than 1 may be used, detection accuracy of 1 is best.

When it is determined from the result of comparison with the criterion Limit that combustion is unstable, the TD of the memory for increasing the amount of fuel is increased.
The value stored in the ML is incremented by a predetermined value DDMLR and reset (steps 20 and 21) to return to the allowable combustion limit. Conversely, if combustion is stable, unnecessary fuel consumption is suppressed by reducing the value stored in TDML by a predetermined value DDMLL (steps 20 and 22). In this way, the target air-fuel ratio on the lean side is corrected. A maximum value and a minimum value are provided for the value of TDML, and the correction range of the target air-fuel ratio is limited below this value. At the time of restart, the initial value (predetermined target value) TDML0 enters TDML.

Finally, in preparation for the next sampling, initial values are stored in Tpmax, Tpmin, Nemax, Nemin and the counter value n in the memory (steps 23 and 24). Note that ∞ in step 23 is the maximum value allowed in the memory.

Here, the operation of this example will be described.

The case where the variables on the horizontal axis shown in FIG. 7 are 1 (there is no load fluctuation), x (there is a small load fluctuation), and y (there is a large load fluctuation) will be described. FIG. 8 shows the relationship between the rotational fluctuation rates again.

In FIG. 8, in a state where there is no load fluctuation (in the case of 1 in the figure), A exceeds the criterion Limit shown by the broken line.
In the range of ~ B, it is determined that there is a rotation fluctuation due to combustion instability, and the fuel is increased. This is the original lean limit control.

However, in the case of x where the load variation is a little even in the same engine, the air-fuel ratio is enriched between C and D at the point C because the determination criterion Limit is reached. The air-fuel ratio is enriched over the entire range.

However, the reason for exceeding the criterion Limit between C and D in the case of x and exceeding the criterion Limit between FG in the case of y is not actually combustion instability, but is influenced by load fluctuation. Therefore, it is not necessary to enrich between C and D and between F and G, and the fuel consumption is degraded by an unnecessary increase in fuel.

On the other hand, the vertical axis is changed from the rotation fluctuation rate to the rotation fluctuation rate / (1 + load fluctuation rate), and is shown in FIG. 9. In FIG. 9, the characteristics in the two cases of x and y with the load fluctuation are shown. Overlap with the characteristic of the case of 1 where there is no load fluctuation. That is,
Since the value when there is a load change under the same air-fuel ratio is the same as the value 1 when there is no load change, the effect of the load change has been eliminated.
When it exceeds it, the rotation fluctuation accompanying the combustion instability always occurs.

As described above, by adopting the rotation fluctuation rate / (1 + load fluctuation rate) as the partner to be compared with the judgment criterion Limit, the accuracy of determination of combustion instability is improved and unnecessary fuel consumption is eliminated.

Further, since sampling is performed only during the steady state, the sampling value is low, and the accuracy of determination of unstable combustion is further improved.

Incidentally, the influence of the load fluctuation on the rotation fluctuation differs depending on the gear position, and the influence is greater at the low speed stage. Therefore, it is necessary to prevent the determination standard from being influenced by the gear position. The reason that the lower gear is more affected is that the greater the gear ratio with respect to the inertia of the vehicle body, the greater the effect of the engine load.

Therefore, in the control unit 2,
According to the result of the gear position determination, the lower the lower gear, the lower the determination criterion Limit.
Is increasing.

FIG. 6 is a flowchart for setting a criterion Limit according to the gear position.
s).

To determine the gear position, the product of the vehicle speed VSP and the gear ratios G _{0 to} G ₄ is compared with the engine speed Ne. Vehicle speed VSP and the product of the gear ratio G ₀ ~G ₄ (engine speed equivalent), since the standards set which ones to use gears which speed region, if G ₀ · VSP ≧ Ne, neutral or the clutch _{OFF, G 0 · VSP> Ne} ≧ G 1
・ In the case of VSP, 1st speed or reverse, G ₁ · VSP>
If Ne ≧ G ₂ · VSP, 2nd speed, G ₂ · VSP> Ne
If ≧ G ₃ · VSP _{3-speed, G 3 · VSP> Ne ≧} G
If it is ₄ · VSP, it is determined that the gear position is in the fourth gear, and if G ₄ · VSP> Ne, it is determined that the gear position is in the fifth gear (steps 32-36).

According to the result of the determination of the gear position, a larger value is set to Limit in the memory at a lower gear (steps 43 to 43).
48). That is, the criterion for the neutral or clutch OFF is a predetermined value Limit0, the criterion for the first speed or the reverse is the predetermined value Limit1,.
If imit5, Limit0>Limit1> ...
> Limit5.

As described above, the lower the gear position is, the higher the rotational fluctuation rate becomes, even with the same air-fuel ratio and the same load fluctuation, and accordingly, the value of Limit is also increased at the same rate in accordance with this. The accuracy of the determination of combustion instability is improved even in a low-speed stage where the influence of is large.

Note that the result of the gear position determination is stored in the GE
AR, and before determining the gear position,
Is transferred to LGEAR of another memory (steps 37 to 42, step 31). Based on the two values GEAR and LGEAR, it is determined that no gear shift has occurred when GEAR = LGEAR, and this determination result is used in step 1 of FIG. 4 described above.

FIG. 10 shows another embodiment corresponding to FIG.

In this example, instead of correcting the rotation fluctuation rate, the criterion Limit is increased in accordance with the load fluctuation rate.

Referring again to FIG. 8 as an example, the criterion Limit
Is raised to the position D when the load variation is a little x, and to the position G when the load variation is large y, thereby eliminating the influence of the load variation. That is, when the rotation fluctuation rate increases due to the load fluctuation, the determination criterion is increased by the same amount as the increase.

As a result, even when there is a load fluctuation, the result of comparison between the rotation fluctuation rate and the judgment criterion is the same as when there is no load fluctuation, and the determination of combustion instability is stabilized as in the previous embodiment.

For this reason, as shown in FIG. 10, the correction rate K of the criterion is looked up from FIG. 11 in accordance with the load fluctuation rate, and this correction rate K is determined based on the criterion when there is no load change (for example, for the 5th speed). The value multiplied by the criterion Limit5) is set as the criterion Limit (steps 51 and 52), and the rotational fluctuation rate ≧ L
If imit, it is determined that combustion is unstable (step 5
3).

On the other hand, in FIG.
Assuming that the value of K at the time of y is Kx and Ky, Kx · Limit5
8, Kx and Ky are determined so that the height of D in FIG. 8 and the height of Ky · Limit 5 are the height of G in FIG. 8, respectively, and the two points thus obtained (two points of H and I in FIG. 11) are determined. If tied, the characteristic of K can be obtained.

Further, in order to make the criterion Limit different according to the gear position and to increase the criterion Limit at a lower gear, it is only necessary to increase the inclination of a straight line at a lower gear as shown in FIG. Regardless of the position, the comparison result between the rotation fluctuation rate and the determination criterion becomes the same, and the accuracy of the determination of combustion instability is improved as in the previous embodiment.

[0071]

According to the first aspect of the present invention, when the condition for operating with the air-fuel ratio leaner than the stoichiometric air-fuel ratio is reached, the actual air-fuel ratio matches the target value with the lean-side air-fuel ratio as the target value. While the feedback control of the amount of fuel supplied to the engine is performed as described above, a criterion for the rotational fluctuation rate is set based on the load fluctuation rate calculated from the detected value of the engine load, and the criterion and the detected value of the engine speed are set based on the criterion and the detected value of the engine speed. The calculated value is compared with the calculated rotation fluctuation rate, and when the rotation fluctuation rate is larger than the determination criterion, the target value is corrected to the side where the air-fuel mixture becomes richer. And wasteful fuel consumption is eliminated.

According to the second aspect of the present invention, when the engine can be operated with an air-fuel ratio leaner than the stoichiometric air-fuel ratio, the engine is set so that the actual air-fuel ratio matches the target value with the lean air-fuel ratio as a target value. While the amount of fuel to be supplied is feedback controlled, the load fluctuation rate is calculated from the detected value of the engine load, and the rotation fluctuation rate is calculated from the detected value of the engine speed, and positive constant values are calculated for the rotation fluctuation rate and the load fluctuation rate. The ratio of the added value is compared with a predetermined criterion, and when the ratio is larger than the criterion, the target value is corrected to the side where the air-fuel mixture becomes rich, so that the same effect as the first invention is obtained. .

According to a third aspect of the present invention, in the first or second aspect, a means is provided for correcting the criterion such that the criterion becomes larger as the gear position becomes lower according to the gear position.
The accuracy of determination of combustion instability is improved even in a low-speed stage where the influence of load fluctuations is large.

In a fourth aspect based on the first or second aspect, the comparison with the determination criterion is performed by using a variation range of the load or a rotation speed.
Means is provided only when the engine is in a steady state in which at least one of the fluctuation widths is equal to or less than a predetermined value , so that the determination accuracy of combustion instability is more stable than in the first or second invention.

[Brief description of the drawings]

FIG. 1 is a diagram corresponding to a claim of the first invention.

FIG. 2 is a diagram corresponding to claims of the second invention.

FIG. 3 is a system diagram of one embodiment.

FIG. 4 is a flowchart for explaining lean limit control.

FIG. 5 is a flowchart for explaining lean limit control.

FIG. 6 is a flowchart for explaining correction according to a gear position as a criterion.

FIG. 7 is a characteristic diagram showing a relationship between (1 + load fluctuation rate) and a rotation fluctuation rate when the air-fuel ratio is constant.

FIG. 8 is a characteristic diagram showing a relationship between a rotation fluctuation rate and an air-fuel ratio.

FIG. 9 is a characteristic diagram showing a relationship between a rotation fluctuation rate / (1 + load fluctuation rate) and an air-fuel ratio.

FIG. 10 is a flowchart for explaining lean limit control of another embodiment.

FIG. 11 is a characteristic diagram of a correction factor K of a criterion of another embodiment.

FIG. 12 is a characteristic diagram of a correction factor K when a gear position is further taken into account in another embodiment.

FIG. 13 is a characteristic diagram showing a relationship between an air-fuel ratio and a rotation fluctuation rate at a constant load in a conventional example.

[Explanation of symbols]

 2 Control unit 3 Injector 4 Air flow meter 7 Crank angle sensor 9 Wide area air-fuel ratio sensor 13 Swirl control valve 32 Air-fuel ratio feedback control means 33 Load fluctuation rate calculation means 34 Judgment reference setting means 35 Rotation fluctuation rate calculation means 36 Comparison means 37 Target value Correction means 41 Ratio calculation means 42 Comparison means 43 Target value correction means

Claims (4)

(57) [Claims] When an operating condition is satisfied at an air-fuel ratio leaner than the stoichiometric air-fuel ratio, the amount of fuel supplied to the engine is set such that the actual air-fuel ratio matches the target value with the lean-side air-fuel ratio as a target value. Means for feedback-controlling the load, and means for calculating a load variation rate from a detected value of the engine load,
A means for setting a criterion for the rotation fluctuation rate based on the load fluctuation rate, a means for calculating a rotation fluctuation rate from a detected value of the engine speed, a means for comparing the rotation fluctuation rate with the criterion, Means for correcting the target value to a side where the air-fuel mixture becomes richer when the rotation fluctuation rate is greater than the determination reference based on the comparison result. 2. Under conditions that allow operation at an air-fuel ratio leaner than the stoichiometric air-fuel ratio, the amount of fuel supplied to the engine such that the actual air-fuel ratio matches the target value with the lean-side air-fuel ratio as a target value. Means for feedback-controlling the load, and means for calculating a load variation rate from a detected value of the engine load,
Means for calculating a rotation fluctuation rate from a detected value of the engine rotation speed; means for calculating a ratio of the rotation fluctuation rate to a value obtained by adding a positive constant value to the load fluctuation rate; An air-fuel ratio control device for a lean burn engine, comprising: means for comparing with a reference; and means for correcting the target value to a side where the air-fuel mixture becomes richer when the ratio is greater than a determination reference based on the comparison result. 3. The air-fuel ratio control device for a lean burn engine according to claim 1, further comprising means for correcting the criterion so that the criterion becomes larger as the speed becomes lower according to the gear position. . 4. The method according to claim 1, wherein the comparison with the criterion is performed by determining a variation range of the load or
The air-fuel ratio control device for a lean burn engine according to claim 1 or 2, further comprising means for limiting the engine to a steady state in which at least one of the fluctuation ranges of the rotation speed is equal to or less than a predetermined value .