JPS63259136A - Air-fuel ratio control device for engine - Google Patents

Abstract

PURPOSE:To maintain an air-fuel ratio in all operational regions at a theoretical air-fuel ratio by providing means for estimating a learning factor in other area of a learning correction factor map according to a deterioration characteristic of a hot wire air flow meter. CONSTITUTION:In an ECU 21 for controlling injectors 8 located in suction passages, basic fuel injection pulse width computing means 22 computes a basic injection pulse width according to outputs from a hot wire air flow meter 11 and a crank angle sensor 18, and feedback correction quantity computing means 24 computes an air-fuel ratio feedback correction quantity according to an output from an exhaust sensor 19. A learning correction factor map 26 sets a learning correction factor for each operational region according to an engine speed and the basic injection pulse width, and learning correction factor computing means 30 updates the correction factor in the map 26 according to an output from the computing means 24. Further, estimated learning correction factor computing means 31 estimates a learning factor in other area of the map 26 from a suction air quantity in a basic area of the map 26 and a correction factor obtained by the computing means 30, according to a deterioration characteristic of the air flow meter 11.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention compensates for measurement errors in the intake air amount due to deterioration of a hot wire air flow meter over time, and makes it possible to obtain the optimum air-fuel ratio in all operating ranges. This invention relates to an air-fuel ratio control device for an engine.

[Problems to be solved by conventional technology and invention] In modern engines equipped with electronically controlled fuel injection systems (EGI), in order to maintain the air-fuel ratio in an optimal state, the basic fuel injection pulse width must first be set. This is then corrected using various correction coefficients to set the actual optimum air-fuel ratio.

The above basic fuel injection pulse width is determined by a function of the engine rotation speed (N) and the amount of intake air (Q). Inexpensive hot wire airflow meters are widely used.

This hot wire airflow meter electrically measures the amount of intake air by using the change in the hot wire resistance value of the hot wire when it is cooled by air flowing through the intake pipe. Since it is constantly exposed in the intake pipe, if it is used for a long time, foreign substances such as fine dust and oil mixed in the intake air will adhere, gradually decreasing the measurement sensitivity and causing so-called aging deterioration. .

As a result, as shown in FIG. 7 (characteristic diagram), in the relationship between the intake air amount Q measured by the hot wire air flow meter and the actual intake air ff1Q', the deterioration characteristic P' of the hot wire air flow meter and the initial When comparing the characteristic P,
While the intake air amount in a certain operating range is actually Qa', the measured intake air amount at the time of deterioration is lower than Qa and the initial state. As a result, the basic fuel injection pulse width is calculated based on the measured intake air ff1Qa that is lower than the actual value, making it difficult to perform appropriate air-fuel ratio control.

In addition, the aging characteristics of the above hot wire air flow meter are as follows:
As shown in the figure, it increases as the intake air mQ moves to a region where it increases. Therefore, the air-fuel ratio control range is limited, and it is difficult to correct leanness in, for example, a high-load operating region.

In addition, recently there are devices that perform learning control on the air-fuel ratio and can instantly determine the optimal air-fuel ratio for each operating range, but if it is possible to perform learning control over the entire operating range, If there are areas where little learning occurs, such as high-load operation areas that are relatively infrequently used, and the hot-wire air flow meter described above deteriorates, the learning coefficient in these infrequently used areas will not be able to cope with this deterioration. , the difference with the required correction coefficient is audible. Therefore, if a learning correction coefficient in a less frequently used range is used during a transient state, the air-fuel ratio will become leaner, leading to engine malfunction and worsening of emissions.

As a means to prevent the hot wire air flow meter from deteriorating over time, for example, Japanese Patent Application Laid-Open No. 56-51670 discloses that after the engine is stopped, a large current is passed through the hot wire, and foreign substances such as dust and oil attached to the hot wire are removed. However, as a result of the hot wire being heated abnormally, the life of the hot wire air flow meter is shortened.

[Object of the Invention] The present invention was made in view of the above circumstances, and the air-fuel ratio characteristics are hardly affected by the aging deterioration of the hot-wire air flow meter, and it can be used not only in areas where learning is frequently performed, but also in areas where learning is frequently performed.
An engine air-fuel ratio that allows a value close to the stoichiometric air-fuel ratio to be obtained even when operating in a region where little learning occurs, and as a result, the stoichiometric air-fuel ratio is guaranteed in the entire operating region, effectively avoiding engine malfunctions and worsening of emissions. The purpose is to provide a control device.

[Means and effects for solving the problems] The engine air-fuel ratio control device according to the present invention adjusts the engine air-fuel ratio to 1l.
The control device includes a calculation unit that calculates the basic fuel injection pulse width based on the intake air m measured by the hot-wire air flow meter and the engine rotation speed measured by the rotation speed sensor, and a a calculation unit that calculates a feedback correction amount of the air-fuel ratio based on the signal; a learning correction coefficient map that sets a learning coefficient for each operating region based on the engine speed and the basic fuel injection pulse width;
1. A learning correction coefficient calculating section that updates the correction coefficient of this learning correction coefficient map based on the output signal from the feedback correction amount calculating section 2), and an actual output fuel injection based on the output signal from each of the above calculating sections. and an output fuel injection pulse width calculation unit that sets a pulse width, the control device is configured to calculate a reference area of the learning correction coefficient map based on a preset deterioration characteristic of the hot wire air flow meter. An estimated learning coefficient calculation unit is provided for estimating learning coefficients in other areas of the learning correction coefficient map based on the intake air amount and the correction coefficient determined by the learning correction coefficient calculation unit.

[Embodiments of the invention] Hereinafter, the present invention will be described in detail with reference to the drawings.

Figures 1 to 6 show an embodiment of the present invention. Figure 1 is a schematic diagram of the main parts of the engine, Figure 2 is a block diagram of electronics and control equipment, and Figure 3 shows the engine rotational speed on the vertical axis. lattice (N lattice),
The horizontal axis is a basic learning correction map showing the basic fuel injection amount grid (Tl grid), Fig. 4 is a flowchart of the air-fuel ratio control device, Fig. 5 is a diagram of aging characteristics of the air flow meter, and Fig. 6 is a graph showing the aging characteristic of the air flow meter.
The figure is a diagram showing the aging characteristics of another air flow meter.

Reference numeral 1 in FIG. 1 denotes an air cleaner. This air cleaner 1 is communicated with a throttle chamber 3 via an intake pipe 2, and this throttle chamber 3 is communicated with an engine main body 5 via an intake manifold 4. Further, a throttle valve 6 is provided within the throttle chamber 3, and a multi-point injector 8 is provided near the intake valve 7 of the intake manifold 4 communicating with each cylinder. In addition, a three-way catalytic converter 1 for purifying exhaust gas is provided in an exhaust pipe 9 communicating with the engine body 1.
0 is inserted.

Further, a hot wire air flow meter 11 is installed immediately downstream of the air cleaner 1. This hot wire air flow meter 11 is formed with a main passage 12 and a bypass passage 13, and a temperature compensation resistor 14 and a hot wire 15 are arranged in this bypass passage 13 in this order from the upstream side.
Both wires 14, 15 form part of a bridge circuit and are connected to a control 16 which detects the air flow rate or mass flow.

The throttle valve 6 also has a throttle opening θ.
A throttle opening sensor 17 is provided on the crankshaft 5a, and a crank angle sensor 1, which is an example of a rotational speed sensor that detects the engine rotational speed N, is installed on the crankshaft 5a.
8 are arranged in a row.

Further, an 02 sensor 19, which is an example of an exhaust sensor for detecting the oxygen concentration in the exhaust gas, faces the exhaust pipe 9, while a cooling water passage 5 formed in the engine body 5 is provided.
The water temperature sensor 20 is facing b.

Furthermore, it is connected to the input side of the control unit 16 of the hot wire air flow meter 11, the throttle opening sensor 17, the crank angle sensor 18, the 02 sensor 19, and the water temperature sensor 20B electronic control unit (ECtJ) 21. ing. Moreover, the above-mentioned engine extractor 8 facing each cylinder is connected to the output side of this E (jJ21).

As shown in FIG. 2, the ECLJ 21 includes the hot wire air flow meter 11 and the crank angle sensor 18.
The basic fuel injection pulse width (
A basic fuel injection pulse width performance unit 22 that determines the basic fuel injection l1a) and an output fuel injection pulse width (output fuel injection amount) calculation unit 2 to which the output signal from this calculation unit 22 is input.
3, a feedback correction amount calculation unit 24 which receives the detection signal from the 02 sensor 19 and calculates a feedback correction amount of the air-fuel ratio, and the crank angle sensor 18,
The detection signals from the throttle opening sensor 17 and the water temperature sensor 20 are inputted to an air-fuel ratio correction coefficient calculating section 25 which calculates a correction coefficient for the air-fuel ratio, the crank angle sensor 18, and the basic fuel injection pulse width ( Basic fuel injection m) An estimated learning determination unit 28 that determines whether the current operating state is in the reference intake air σ region based on the output signal from the calculation unit 22, and determines whether or not to perform estimated learning based on the determination. , a P-P value (maximum value - minimum value) detection section 29 that detects the maximum value and minimum value of the feedback correction amount of the air-fuel ratio from the output signals of the 02 sensor 19 and the feedback corrector calculation section 24; P- from the P-P value detection section 29 based on the output signal from the estimated learning determination section 28
A learning correction coefficient calculation unit 30 that calculates a learning correction coefficient from Phi. and an estimated learning correction coefficient calculating section 31 that estimates learning correction coefficients for all operating regions of the intake air barrier larger than the reference intake air amount region based on the output signal from the estimated learning determining section 28.

Further, the learning correction coefficient calculation section 30 and the estimated learning correction coefficient calculation section 31 are connected to the learning correction coefficient map 26 via a path line.

As shown in FIG. 3, this learning correction coefficient map 26 is a two-dimensional map, in which the engine speed (N) and the basic fuel injection pulse width are adjusted so that the diagonal direction of 45 degrees to the right is a constant intake air amount. (Tp) lattice (N lattice, T
p lattice) are arranged in 16×16 grids in the figure.

In the learning correction coefficient map 26 shown in the figure, Tpo lattice and N
The area shown by the hatching connecting the O grid is the standard intake air m
This reference intake air m-region MPo indicates the entire diagonal direction of the most frequently used area that is arbitrarily selected from among the learning areas of the learning correction coefficient map 26. In addition, this reference intake air m area MP
The entire region of intake air m larger than o (lower right of the hatched part in the figure) becomes the deterioration estimation learning region.

Further, this learning correction coefficient map 26 is connected to the output fuel injection pulse width (output fuel injection ff1) calculation section 23 via a pass line. Further, the output fuel injection pulse width (output lJ fuel injection ff1) calculation unit 23 receives the feedback correction lowest nm 24 and the output signal from the air-fuel ratio correction coefficient calculation unit 25, and outputs the output fuel injection pulse width ( output fuel injection amount) is determined, and a signal is output to the injector drive section 27, and this injector drive section 2
7 outputs a fuel injection signal to the injector 8 provided in each cylinder.

Next, the control operation of the air-fuel ratio control device with the above configuration will be explained.

While the engine is running, fresh air flows in from air cleaner 1,
The air flows through the main passage 12 of the hot-wire airflow meter 11, the intake pipe 2, the throttle chamber 3, and the intake manifold 4, and is sucked into the combustion chamber. Furthermore, combustion gas passes through the exhaust pipe 9, passes through the catalytic converter 10, and is released into the atmosphere from a muffler (not shown).

A part of the intake air flowing in from the air cleaner 1 flows through the bypass passage 13 of the air flow meter 11, contacts the temperature compensation resistor 14 facing the bypass passage 13, and cools the hot wire 15. do. The control unit 16 of the air flow meter 11 is provided with two reference resistors that together with the temperature compensation resistor 14 and the hot wire 15 constitute a bridge circuit.
The hot wire resistance value of the hot wire 15, which changes as it is cooled by the intake air, is compensated for by the temperature compensating resistor 14, and the air flow rate, that is, the mass flow rate, is electrically detected, and the air is sucked into the electronic control unit (ECU>21). Output as air amount signal Q].

The ECtJ21 also includes an engine rotational speed signal N1 detected by the crank angle sensor 18, a throttle opening signal θ detected by the throttle opening sensor 17,
A cold 1 water temperature signal Twq from the water temperature sensor 20 and an oxygen concentration signal VO2 from the 02 sensor 19 are input simultaneously.

Then, the intake air amount signal Q detected from the air flow meter 11 and the engine rotation speed signal N detected by the crank angle sensor 18 are input to the basic fuel injection pulse width (basic fuel injection amount) calculation section 22. The basic fuel injection pulse width (basic fuel injection amount) is determined.

That is, if the basic fuel injection pulse width (basic fuel injection amount) is To, then To-k・(Q/N) where: Constant Q: Intake air amount measured by air flow meter 11 N: Crank angle sensor 18 It is determined by the engine rotation speed measured by .

On the other hand, the engine rotational speed signal N1 from the crank angle sensor 18 and the basic fuel injection pulse width calculated by the basic fuel injection pulse width (basic fuel injection m) calculation unit 22 are estimated by the estimated science judgment unit 28. is input.

In this estimation west determination unit 28, the engine rotation speed signal N
The information on the basic fuel injection pulse width signal Tp is compared with the learning correction coefficient map 26 shown in FIG. 3 to determine whether the current operating state is within the reference intake air amount region MPO.

As a result of the determination, the current operating state is within the above reference intake air amount region MP.
If it is determined that the value is other than o, the learning correction coefficient calculation unit 3
A processing command signal is output to 0, and a non-processing command signal is output to the estimated learning correction coefficient calculating section 31.

In the learning correction coefficient calculating section 30, first, the P-P value (R Otake-minimum value) detecting section 29 detects that while the output signal VO2 from the 02 sensor 9 passes through the slice level a predetermined number of times,
Feedback correction? 11 The maximum value a and the minimum value of the air-fuel ratio feedback correction amount α output from the bright section 24 are expressed as
The P value detection unit 29 detects the a-b value, calculates the learning correction coefficient of the current learning area based on this a-b value, and rewrites the area in the learning correction coefficient map 26.

That is, first, the intermediate value λ between the maximum value a and the minimum value S is calculated.

λyoa+b Next, the current correction coefficient is determined based on the intermediate value λ.

In other words, the initial data stored in the area is
If the latest data is written as X5, then ×”-X+M-Δ
λ where M: a constant determined by Aλ 0≦M<1 Δλ: determined by the amount of deviation of the intermediate value λ from the control center.

As a result, this value X is the current operating range hit (for example, N
o , Tpo) becomes the learning correction coefficient Ko (X'=K
o).

Further, if the estimated learning determining section 28 determines that the current driving condition is within the reference intake air amount region MPo, a processing command signal is sent to the learning correction coefficient calculation section 30 and the estimated learning correction coefficient vXX group section 1. is output.

As a result, the learning correction coefficient calculation unit 30 calculates the learning correction coefficient Ko by the same calculation as described above, and the estimated learning correction coefficient calculation unit 31 calculates the reference intake air m region MPo based on the learning correction coefficient Ko. A learning correction coefficient K for the entire region of larger intake air m is estimated, and the learning correction coefficient map 26 is rewritten using this estimated learning correction coefficient K.

First, when calculating this estimated learning correction coefficient, the value of the measured intake air 51Qo may be different from the actual intake air ff1Qo due to aging deterioration of the hot wire air flow meter 11 as described above.
Since the measured value is smaller than ', the deterioration characteristics of this hot wire air flow meter 11 are first patterned.

Figure 5 shows a pattern of the deterioration characteristics of a certain hot-wire air flow meter 11 determined through experiments.If the deterioration rate is Y, then Y=aQ2...■ a: A constant related to operating time. shown.

In other words, although the progress of deterioration of the hot-wire air flow meter 11 varies individually, it is considered to be a function of operating time, so it is possible to pattern it. If the deterioration rate in MPo in Figure 3 can be determined, other areas can be determined.
In particular, it is possible to estimate the deterioration rate in a region where the amount of intake air is large, where the deterioration rate is high.

When the above equation 0 is applied to the measured intake air ff1Qo in the reference intake air open area MPo in FIG. 3 (see FIG. 5), the deterioration rate Yo here is Yo −aQo 2
・・・・・・■ becomes. .

Also, the actual intake air ff1Qo' at this time is calculated from the above equation 0 as follows: Qo' - Qo / (1-Yo) ・
”■, and the above learning correction coefficient Ko in this region is Ko
-Qo' /Qo -1/ 1-Yo = ■, so the coefficient a of the deterioration rate can be found from the above equations 0 and 0, and is a-Yo/Qo2- (1-1/Ko)/Qo2...
...■ From this, the deterioration rate in the area of measured intake air fnQ is '
, Y=aQ2-02・(1-1/Ko)/Qo2・
...■ Therefore, the estimated correction learning coefficient for the measured intake air IQ in each region is: K=1/<1-Y)=1/N-(Q2 ・(1-
1 /Ko) /Qo 2)]
--■, and the learning correction coefficient calculation section 3
Based on the learning correction coefficient KO of the reference intake air region MPo calculated as 0, the estimated learning correction coefficient K of each intake air region M is determined according to the deterioration pattern of the hot wire air flow meter 11 (see FIG. 5).

In addition, in the estimated learning area, at the time of estimated learning calculation,
Estimation learning may not be performed for areas where letters have already been erased a predetermined number of times or more.

Then, based on the estimated learning correction coefficient K obtained by the estimated learning correction coefficient calculating section 31, the learning correction coefficient map 26 is
All of the intake air m regions that are larger than the reference intake air amount region MPo are rewritten.

The deterioration characteristics of the hot wire air flow meter 11 are as follows:
For example, it may be divided into patterns as shown in FIG. 6, and the estimated learning correction coefficient in this case is determined as a function of the deterioration rate Y determined for each section and the measured intake air amount Q.

Then, the output fuel injection pulse width (output fuel injection amount) calculation section 23 receives the signal TI) outputted from the basic fuel injection amount pulse width (basic fuel injection amount) calculation section 22 and the signal TI output from the feedback correction calculation section 24. The output feedback correction signal α, the signal θ from the throttle level sensor 17, the signal N1 from the crank angle sensor 18, and the water temperature sensor 2.
The learning correction coefficient signal obtained from the learning correction coefficient map 26 is input together with the air-fuel ratio correction coefficient signal calculated in the air-fuel ratio correction coefficient calculating section 25 based on the signal Tw from 0, and the optimal output fuel injection pulse is input. The width (output fuel injection ta>Tr) is determined and output to the drive unit 27, and the drive unit 27 outputs a fuel injection signal to each injector 8.

The procedure for rewriting the learning correction coefficient 26 will be explained according to the flowchart shown in FIG.

First, in step 5101, the hot wire air flow meter 11
The basic fuel injection pulse width (basic fuel injection Djl>To is calculated based on the intake air ff1Q from the engine and the engine rotation speed N from the crank angle sensor 18. Then, step 510
Proceeding to step 2, the current operating state is the standard intake air amount region MP based on the basic fuel injection pulse width (basic fuel injection amount) TI) and the engine rotation speed N.

It is determined whether the intake air amount is within the reference intake air amount region MPo, and if it is outside the reference intake air amount region MPo, the process advances to step 5103, and if it is determined that it is within the reference intake air amount region MPo, the flow advances to step 5104.

Then, proceeding to step 5103, the intermediate value λ between the maximum value and the minimum value of the feedback correction value α calculated from the output voltage 02 from the 02 sensor 19 is calculated, and the learning correction coefficient Ko is determined based on this.

Then, the process proceeds to step 3106, where the learning correction coefficient K
The learning correction coefficient map 26 is rewritten according to o.

Also, from step 3102 to step 5104
When the process proceeds to step 3103, the learning correction coefficient KO for the region MPo is determined by the same means as in step 5103, and then the process proceeds to step 3105.

Then, in step 5105, the measured intake air ff1Q in the reference intake air amount region MPo (see FIG. 3) is
o and the learning correction coefficient KO, all estimated learning correction coefficients K in the region where the intake air amount is larger than the reference intake air amount region MPo are determined, and the process proceeds to step 8106.

As a result of the above, even if the hot wire air flow meter 11 deteriorates over time and the intake air ff1Qo is measured at a lower value than the actual intake air IQo', it is compensated for and maintains the optimum air-fuel ratio characteristic in all operating ranges. can be obtained.

Note that the air-fuel ratio control device according to the present invention is not limited to the above-described embodiment, and may also perform estimation learning on the side where the intake air m is smaller than the reference intake air amount region MPo at the same time.

[Effects of the Invention] As explained above, according to the present invention, the control device that controls the air-fuel ratio of the engine is configured to calculate the intake air in the reference region of the learning correction coefficient map based on the deterioration characteristics of the hot-wire air flow meter set in advance. Since an estimated learning coefficient calculation unit is provided that estimates learning coefficients in other areas of the learning correction coefficient map using the correction coefficients determined by the correction coefficient calculation unit, the reference intake air amount area When learning, almost all other operating ranges are estimated and learned, and all operating ranges can follow the progress of deterioration of the hot wire air flow meter, and even in areas where actual training is rarely performed, the theoretical empty is always maintained. The fuel ratio can be maintained, engine malfunctions and emissions deterioration can be effectively prevented, and comfortable drivability can be obtained.

[Brief explanation of drawings]

1 to 6 show one embodiment of the present invention, FIG. 1 is a schematic diagram of the main parts of the engine, FIG. 2 is a block diagram of the electronic control device, and FIG. 3 is an engine rotational speed grid on the vertical axis. (N lattice), the horizontal axis does not show the basic fuel injection amount lattice (Tl lattice), the basic learning correction map, Fig. 4 is a flowchart of the air-fuel ratio control device, Fig. 5 is a diagram of aging characteristics of the air flow meter, FIG. 6 is a characteristic diagram of aging deterioration of another air flow meter, and FIG. 7 is a characteristic diagram of the actual intake air amount and measured intake air amount in the initial state and the deteriorated state of the air flow meter. 11... Hot wire air flow meter, 18... Rotation speed sensor (crank angle sensor), 19... Exhaust sensor (
02 sensor), 21... Control device, 22... Basic fuel injection pulse width calculation section, 23... Output fuel injection pulse width calculation section, 24... Feedback correction performance cylinder section, 2
6...Learning correction coefficient map, 30...Learning correction coefficient map, 31...Estimated learning correction coefficient calculation unit, Ml). ...Reference area, N...Engine speed, Q...Intake air amount, Tp...Basic fuel injection pulse width, Y...
Deterioration characteristics. Fig. 3 Fig. 4 Fig. 5 Fig. 7 Increase 11 Mugi Inhalation Genkido

Claims (1)

[Claims] The control device that controls the air-fuel ratio of the engine includes a calculation unit that calculates the basic fuel injection pulse width based on the intake air amount measured by the hot wire air flow meter and the engine rotation speed measured by the rotation speed sensor. , a calculation unit that calculates a feedback correction amount for the air-fuel ratio based on a signal from the exhaust sensor, and a learning correction coefficient map that sets a learning coefficient for each operating region based on the engine speed and the basic fuel injection pulse width. and a learning correction coefficient calculation unit that updates the correction coefficient of this learning correction coefficient map based on the output signal from the feedback correction amount calculation unit, and a learning correction coefficient calculation unit that updates the actual output fuel injection pulse width based on the output signal from each of the calculation units. In the air-fuel ratio control device for an engine, the control device includes a reference for the learning correction coefficient map based on a preset deterioration characteristic of the hot-wire air flow meter. Based on the intake air amount in the region and the correction coefficient obtained by the learning correction coefficient calculation section,
An air-fuel ratio control device for an engine, characterized in that an estimated learning coefficient calculating section is provided for estimating learning coefficients in other areas of the learning correction coefficient map.