JP2514627B2 - Air-fuel ratio control device for internal combustion engine - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an air-fuel ratio control device for an internal combustion engine, and more particularly to improving the air-fuel ratio control accuracy in a transient operation state of a spark ignition type internal combustion engine. To an improved air-fuel ratio control device.

(Prior Art) In an internal combustion engine for vehicles, etc., how to appropriately control the amount of fuel or the air-fuel ratio supplied to the engine from the viewpoint of meeting the demand for exhaust gas purification while improving the output performance and drivability originally required for the engine. Whether you do it is an important issue. In particular, vehicle engines are used in a wide operating range from low-speed low-load to high-speed high-load, so the suitability of air-fuel ratio control in transient operating conditions such as acceleration and deceleration greatly affects drivability and exhaust emissions. .

Therefore, based on an electronically controlled fuel injection device with excellent fuel metering accuracy, the fuel injection amount is increased or decreased during acceleration or deceleration so that an appropriate air-fuel ratio can be obtained in all operating conditions including transitions. The control device or control method described above is being adopted in many vehicle engines. (As a known example of this type of control method, for example, JP-A-58-144632, 144634, 144636,
See 150033, 150042 and 150043. ) The reason why such transient correction is necessary is that fuel that adheres to the inner wall surface of the intake pipe or the intake port before reaching the engine cylinder or fuel that is not sucked and floats in the intake pipe (these fuels are The amount of "intake system adhesion, floating fuel") has an effect on the air-fuel ratio or engine performance during a transient state. For example, if only an amount of fuel proportional to the intake amount is supplied during acceleration, the Since the part adheres to the intake system and causes a delay in the supply response, the actual air-fuel ratio becomes too thin and the acceleration performance deteriorates.

(Problems to be solved by the invention) By the way, the adhesion of the intake system and the amount of floating fuel change according to the operating state of the engine, and the rotation speed and the engine temperature as well as the absolute pressure of the intake pipe and the volatilization of the fuel. However, in the conventional air-fuel ratio control, the transient fuel excess / deficiency amount is approximately calculated by a correction method that has been experimentally determined in advance using the change of the intake pipe pressure as a parameter. It is based on the method of optimizing the air-fuel ratio by making a correction according to the cooling water temperature.Therefore, as described above, the intake system always fluctuates based on various factors It is not always possible to obtain an appropriate air-fuel ratio, and it is inevitable that an error will occur except under the specific operating conditions that are the design points.

However, in order to solve this, it is necessary to detect and correct all the factors that affect the adhesion of the intake system and the amount of floating fuel, but in this case there are many judgment conditions regarding the necessity of correction, etc. However, many steps are required for the matching work to satisfy the requirements of drivability and exhaust emission.

Therefore, paying attention to these points, the intake system adhesion and the floating fuel equilibrium amount M0 are calculated, and the difference value M0-M between the equilibrium amount M0 and the intake system adherence and floating fuel prediction variable M at that time point. And a correction coefficient DK indicating how much this difference value is reflected in the correction of the fuel injection amount, the transient correction amount DM is obtained, and the predictive variable M is updated in synchronization with the fuel injection. A person has previously proposed it (see Japanese Patent Application No. 60-243605), and the present invention is an improvement over the previous application.

In other words, in the prior application, instead of detecting the factors related to the intake system adhesion and the floating fuel amount, the intake system adhesion and the floating fuel amount were directly dealt with. It has become possible to obtain an air-fuel ratio characteristic with good responsiveness.

However, the basic pulse width Tp adopted in the L-Jetronic system (for example, in the case of measuring the intake air flow rate Qa, Tp = K · Qa / N, where N is the rotation speed,
K is a constant. ) Also fluctuates even in the steady state of the operating state, the final fuel injection pulse width TI calculated on the basis of this Tp fluctuates, which may cause torque fluctuations. This is not a preferable phenomenon because it induces a phenomenon. In particular, when the present apparatus is applied to a lean-burn engine, such an engine is easily affected by fluctuations in pulse width, so fluctuations in pulse width must be taken into consideration.

Explaining the cause of such fluctuations in the pulse width, TI is calculated from the following equation.

TI = (Tp + DM) × β × COEF + Ts where DM = DK ・ (M0-M); excess / deficiency M0 = f (Tp, N, Tw); equilibrium amount DK; correction coefficient Tw; cooling water temperature β; air-fuel ratio feedback Correction coefficient COEF; Sum of various correction coefficients Ts; Invalid pulse width Here, Tp fluctuates even in the steady state because the signal of the air flow sensor or the intake pipe pressure sensor fluctuates due to intake pulsation in the intake system. This is because Tp, which is a function of flow rate or intake pipe pressure, also fluctuates.

When Tp fluctuates, TI also fluctuates. However, since the equilibrium amount M0, which is a characteristic part of the prior application, is calculated using Tp, M0, and therefore DM, fluctuates due to Tp fluctuation. That is, in the previous application, TI is
Since it is given as the sum of DM, if simply considered, it can be said that the fluctuation of Tp is doubled to affect TI.

In order to suppress such fluctuations in TI, it is possible to deal with it by performing signal processing such as smoothing the fluctuating M0 signal, but if smoothing is performed transiently, the responsiveness will be impaired. It is difficult to determine the degree.

Therefore, it is conceivable to use the throttle valve opening signal as the engine load signal that is not affected by the intake pulsation. However, as the load signal, originally, good linearity with respect to the engine load is required. The opening α is inferior to the intake pipe pressure, the air flow rate, etc., which are affected by the intake pulsation. Therefore, a smooth characteristic curve cannot be obtained by creating the M0 map based on α. In addition, the characteristic of the throttle valve opening α has individuality because it varies depending on the shape of the throttle valve and the design structure of the intake pipe passage around it, and therefore, the matching work of M0 experimentally determined is limited. This must be performed for each valve, which makes matching work difficult.

Further, even if the linearity as the load signal is shelved, calculating M0 using the throttle valve opening α itself causes another problem.

For example, at the reference temperature Tw1 of the cooling water temperature, when the map search is performed using α itself and the engine rotation speed N as parameters, the contents of the M0 map (M01 map) are shown in FIG. In the area (the area surrounded by the broken line), the value of M01 changes abruptly even if there is a slight change in α, so even if you try to obtain the same level of control accuracy as in other areas, the grid points You have to be close. In that case, a large amount of data is required to fill the grid points, and the resolution of the values of the grid points is insufficient, for example, 8 bits, and the calculation becomes complicated. In the following description, the case where there are two parameters (α and N) as shown in FIG. 15 is called a map, and the case where there is one parameter is called a table.

As a result, the load signal is not affected by the intake pulsation and is required to have good linearity with respect to the load.
In addition, it is required that the sudden change region does not occur.

The present invention has been made in view of these problems, and the minimum opening area (hereinafter simply referred to as "opening area") that is uniquely determined according to the throttle valve opening α and is formed between the surrounding intake passages. The value obtained by dividing A by the engine rotation speed N.
A / N is adopted as the basic value for control, and this A / N value and other 1
Intake system adhesion and floating fuel equilibrium amount M0 corresponding to one operating state signal are stored in advance as a map value, and the map can be searched from the A / N value at that time and another operating state signal. The purpose is to improve the prior application by obtaining the equilibrium amount M0.

(Means for Solving the Problems) In order to achieve the above object, according to the present invention, as shown in FIG. 1 or FIG. 16, the operating state of the engine is set to at least the engine speed, the intake air amount, and the engine temperature. Including an operating state detection means 1, a basic injection amount calculation means 2 for calculating a basic injection amount Tp of fuel based on the engine speed and an intake air amount, and an intake throttle valve opening signal The opening area calculating means 3 for calculating the opening area A, the dividing means 4 for dividing the opening area A by the engine speed N, the value A / N obtained by dividing the opening area A by the engine speed N, etc. Of the intake system corresponding to one operating state signal (for example, engine temperature or engine speed), and the equilibrium amount storage means 5A and the division means 4 for preliminarily storing the equilibrium amount M0 of the floating fuel as a map value. Divided value A / N and one other operating state Deposition of an intake system corresponding to the item, the equilibrium amount calculation means 5 for the equilibrium amount M0 of suspended fuel consisting of the equilibrium amount reading means 5B for reading from the equilibrium value memory 5A (see FIG. 1) or the opening area A
Of the intake system corresponding to the value A / N divided by the engine speed N, the basic equilibrium amount storage means 5C for pre-storing the basic equilibrium amount M0AN of floating fuel as a table value, the division value obtained by the division means 4 The basic equilibrium amount M0AN corresponding to the A / N is read from the basic equilibrium amount storage means 5C, and the basic equilibrium amount detection means 5D and the correction amount for another one operation state signal (for example, engine speed) are stored in advance as table values. Correction amount storage means 5
E, a correction amount reading unit 5F that reads a correction amount corresponding to another one operation state signal from the correction amount storage unit 5E, and a value obtained by correcting the read basic equilibrium amount M0AN with the read correction amount. An equilibrium amount calculation means 5 (see FIG. 16) including a correction means 5G for setting the equilibrium amount M0 of adhered and floating fuel,
Equilibrium amount M0 of adhered and floating fuel calculated by the equilibrium amount calculation means 5
And a difference value calculation means 6 for calculating a difference value M0-M between the adhesion of the intake system at that time and a predictive variable M of the floating pigment, and a difference value M0-M calculated by the difference value calculation means 6 for the fuel injection amount. Correction coefficient DK indicating how much the correction coefficient DK is reflected on the basis of the engine speed, engine load and engine temperature, and based on the difference value M0-M and the correction coefficient DK. The transient correction amount calculating means 8 for calculating the transient correction value DM, and the transient correction amount DM calculated by the transient correction amount calculating means 8 and the predictive variable M of the adhered and floating fuel are added in synchronization with the fuel injection, A prediction variable calculation means 9 for updating the prediction variable M with the added value,
Fuel injection amount calculation means for calculating the fuel injection amount Ti based on the basic injection amount Tp calculated by the basic injection amount means 2 and the transient correction amount DM calculated by the transient correction amount calculation means 8 to output an injection signal. A fuel supply means 11 for supplying fuel to the engine based on the injection signal is provided.

(Function) Since the equilibrium amount M0 is calculated using the divided value A / N that is not affected by the intake pulsation as a parameter, it is possible to avoid the equilibrium amount M0 from varying with the intake pulsation, and as a result, the torque variation Since it becomes smaller, the rattling vibration of the vehicle becomes smaller.

Further, since the division value A / N and the equilibrium amount M0 corresponding to another one of the operating state signals are stored in advance as map values, the number of map grid points is set to be almost flat without a sudden change region. You don't have to be close. As a result,
The amount of map data is small, the resolution of map grid points is only 8 bits, and the software and control unit capacity are small.

In addition, the basic equilibrium amount corresponding to the divided value A / N and the correction amount corresponding to one other operating state signal are stored in advance as respective table values, and the divided value A / N at that time and the other 1 The fluctuation of the equilibrium amount M0 can also be reduced by searching the corresponding table values from the one operating state signal to obtain the basic equilibrium amount and the correction amount, and setting the value obtained by correcting the basic equilibrium amount by the correction amount as the equilibrium amount M0. At the same time, the amount of data in the table is small, and the time required to calculate the equilibrium amount M0 is greatly shortened, improving responsiveness.

 Hereinafter, a specific description will be given using examples.

(Embodiment) FIG. 2 is a mechanical configuration of an embodiment in which the present invention is applied to a so-called single-point injection type electronically controlled fuel injection device in which one fuel injection valve 17 is provided in an intake passage 16 upstream of a throttle valve 15. Is represented.

Explaining from the part that is almost the same as the previous application, the intake flow rate
From the air flow rate sensor 20 that detects Qa, the crank angle sensor 21 that detects the engine speed N, the water temperature sensor 22 that detects the cooling water temperature Tw, and the air-fuel ratio sensor 23 that detects the actual air-fuel ratio necessary for feedback control. Various signals of are input to the control unit 30
At 30, the drive control of the injection valve 17 is performed based on these signals.

In contrast to such a configuration, in the present invention, it is necessary to know the opening area A of the intake air, and the opening area A is uniquely determined if the intake throttle valve opening α is known. Therefore, the opening sensor 25 that detects the throttle valve opening α is provided. In addition, 26 is a neutral switch, 27 is a clutch switch, and 28 is an idle control valve.

The control unit 30 is composed of a microcomputer including a CPU, a RAM, a ROM, an I / O device, etc., has all the functions of each means 2 to 10 shown in FIGS. 1 and 16, and controls the air-fuel ratio ( Injection amount control) is centralized.

By keeping the fuel pressure to the injection valve 17 constant, the injection amount is proportional to the valve opening pulse width, so that the valve opening pulse width is actually calculated in the control unit. Will be described as pulse width control.

3 to 7 are flow charts for explaining the routine executed in the control unit. Of these, FIG.
Fig. 4 shows the main routine of pulse width control,
FIG. 7 to FIG. 7 correspond to a subroutine for obtaining a correction value or the like used to that extent. The numbers in the figure indicate the process numbers. The processes shown in FIGS. 3, 5 to 7 and 12 are performed every predetermined time or in synchronization with the engine rotation, and only the process shown in FIG. 4 is performed in the engine rotation. It is executed in synchronization (to be exact, in synchronization with injection).

The basic pulse width control of the injection valve is well known. For example, as shown in FIGS. 3 and 4, the intake air flow rate Qa and the rotation detected by the air flow rate sensor 20 and the crank angle sensor 21, respectively. A basic pulse width Tp (= K · Qa) with which a predetermined air-fuel ratio (for example, a theoretical air-fuel ratio) can be obtained using the speed N
/ N, where K is a constant. ) Is obtained, and this is multiplied by the feedback correction coefficient β determined based on the output of the air-fuel ratio sensor 23 and the total COEF of other correction coefficients, and the invalid pulse width Ts is further added to obtain the final injection pulse width TI (= Tp
・ COEF ・ β + Ts) is obtained and a drive signal based on this TI is given to the injection valve 17 (40, 51, 52).

In the prior application, in the process of obtaining the injection pulse width TI in this way, a correction corresponding to a more transient operating state is made by paying attention to intake system adhesion and floating fuel. 44 (specifically, 42 is a balance amount calculating means, 43 is a correction coefficient calculating means, and 44 is a portion functioning as a transient correction amount calculating means), 50, 53 (50 is a fuel injection amount calculating means) in FIG. , 53 is a part that functions as a predictor variable calculation means.).

The features similar to those of the prior application will be described later, and the characteristic part of the present invention will be described. This is because, in the prior application, the adhering amount of the intake system and the equilibrium amount M0 of the floating fuel, which are the basis of the correction, are calculated using the three parameters N, Tp, and Tw in the previous application. Affected Tp
Instead of the opening area A and the engine speed N divided by A /
Using N, the equilibrium amount M0 corresponding to this division value A / N is stored in advance as a map value, and this map value is searched from the division value A / N at that time to find the equilibrium amount M0. .

That is, in FIG. 3, step 41 for calculating A / N is newly introduced. The A / N calculation is executed by the subroutine shown in FIG. 5, and in the figure, steps 60 and 61 correspond to the functional portions of the means 3 and 4 shown in FIGS. 1 and 16, respectively. Equivalent to. First, the opening area A is calculated from the throttle valve opening α. This is obtained by searching the table having the contents shown in FIG. 8, and the obtained A is divided by the engine rotation speed N. , A / N is calculated.

The characteristics shown in FIG. 8 are those when the throttle valve 15 is a butterfly type. Therefore, in the case of a shutter type or the like, the characteristics are also different from those in FIG.

Next, the map search of M0 based on this A / N is performed by the subroutine shown in FIG. 6, and basically the same as the previous application, but the map used in this routine is the previous application. It makes a big difference.

Therefore, the contents of this map are shown in FIG. 9, which shows the characteristics of the map of M0 (M0 map) with respect to N and A / N at the reference temperature Tw1 of the cooling water temperature. A flat characteristic curve is obtained and no abrupt change region is generated. The fully open line in the figure is the line when the maximum opening area Amax is reached.

Next, the operation of calculating M0 using such a map will be described. A is the throttle valve 15 and the intake passage 16 around it.
Since it is a value that is uniquely determined according to the shapes of both, it does not change due to the influence of intake pulsation. Also,
Since the opening area A has a correlation with the intake pipe pressure and the air flow rate, the linearity with respect to the load is also good.

Therefore, the factor A that is not affected by the intake pulsation
Since M0 is calculated using as a parameter, the influence of intake pulsation can be eliminated from the DM calculated using this M0. As a result, it is possible to eliminate the fluctuation due to inspiration pulsation from one of the two factors affecting TI, that is, DM, so that the effect of Tp is not doubled, and therefore TI is only affected by the effect of Tp fluctuation. It's done.

However, in the prior application, since the two fluctuation factors Tp and DM are used as they are for the calculation of TI, the fluctuation of Tp is almost doubled and contributes to TI, which causes the torque fluctuation due to the fluctuation of the injection amount. It is considered that a so-called jerky phenomenon occurs.

Therefore, in the present invention, in addition to the effect of the prior application, it is possible to avoid the fluctuation of the equilibrium amount M0 due to the intake pulsation, and as a result, the torque fluctuation can be reduced, so that the rattling vibration of the vehicle can be reduced.

Also, if M0 corresponding to A / N and N is used as the map value, as shown in FIG. 9, there is no sudden change region where the characteristic curve changes rapidly, and it is almost flat. Moreover, it is not necessary to make the number of map grid points dense. As a result, the amount of map data is small, the resolution of the map grid points is only 8 bits, and the software and control unit capacity are small.

On the other hand, if a map of M0 is created using A itself, a sudden change region will be generated as in the case shown in FIG. 15, and if measures are taken to make the grid points dense in this sudden change region. If you do, you need a lot of data, 8
If the bit resolution is insufficient, or if nothing is done as it is, the control accuracy will be reduced and it will not be possible to comply with the actual change in the operating state.

Next, A / N is used in place of Tp to outline other functions, which are performed in the same manner as the previous application.

In step 42 of FIG. 3, the adhering amount of the intake system and the equilibrium amount M0 of the floating fuel are calculated using the three parameters N, A / N and Tw.
As shown in FIG. 6, this is obtained by linear approximation interpolation calculation processing using map search values. Specifically, five different preset reference temperatures Tw0 to Tw4 (Tw0>...> Tw4) that are preset within the range of the temperature change width that the cooling water temperature Tw can actually take.
For convenience, five maps are provided by actual measurement in which M00 to M04 (M0 for Tw0 to Tw4) are assigned with A / N and N as parameters. Of these, the content of the map that gives M01 (M01 map) is shown in FIG. 9, and it was described above that this is a characteristic part of the present application. Then, assuming that the actual water temperature Tw is Tw ≧ Tw1, the N, A at that time for each of the reference temperatures Tw0, Tw1 above and below Tw.
Map search is performed according to / N to obtain map values M00 and M01. These values M00, M01 and temperature difference (Tw0-Tw), (Tw0
-Tw1) performs linear approximation interpolation calculation to finally determine M0 (steps 70 to 73).

Next, in step 43 of FIG.
Coefficient DK that indicates the rate at which the predicted value (predicted variable) M of the intake system adhesion and floating fuel at present approaches M0 per unit cycle (for example, each crankshaft revolution) Is calculated from the product of the coefficients DKTw and DKN (steps 90 to 92 in FIG. 7).

The DKTw is based on the excess / deficiency amount (transient correction amount) DM and the water temperature Tw per unit cycle obtained in the previous process, and the
It is a value obtained by searching the map formed as shown in the figure. For example, the larger the excess / deficiency amount DM, the larger the value is set to eliminate the excess / deficiency amount quickly. Also, DKN
Is a value obtained by searching a map similarly formed as shown in FIG. 10 based on N and Tp, and is set to be larger as the rotational speed becomes smaller. DK may be a constant value or a simple function, but here, N, A / N, DM, and Tw are experimentally obtained as parameters, and highly accurate correction is possible.

In step 44, the coefficient DM is multiplied by the difference between M0 and the predicted value M to calculate the excess / deficiency amount DM (=
Calculate DK (M0-M)). Here, the predicted value M is the predicted value of the adherence of the intake system and the floating fuel at that time, and therefore (M0-M) indicates the excess / deficiency amount from the equilibrium amount, and this value (M0-M) is It is further corrected by the coefficient DK obtained by using N, Tp, DM and Tw as parameters.

FIG. 4 shows the processing for calculating the final fuel injection pulse width TI by adding the excess / deficiency amount DM thus obtained. In step 50, the basic pulse width Tp is corrected and the basic fuel pulse width Tp is corrected. TpF (= Tp + DM) is required. Then, in the prior application, this TpF is the conventional T
Although it is replaced by p, in the present invention, the influence of the intake pulsation is eliminated from the second term DM of the two terms constituting this TpF. The DM may be reduced by a predetermined amount during deceleration. In this case, when a highly volatile fuel is used, it is possible to prevent the air-fuel mixture from being temporarily diluted in the initial stage of deceleration.

Finally, in step 53, the next predicted value M (= old M + DM) is calculated by adding the current excess / deficiency amount DM to the previous predicted value M (represented as “old M”) for the next processing. In the process of FIG. 4, for example, TI is calculated and injected for each revolution of the engine crankshaft, and the predicted value M is updated each time.

Next, FIG. 12 is a flow chart for explaining a routine for calculating M01 of the second embodiment of the present invention.

The difference from the first embodiment is that, in the first embodiment, M01 is obtained at one time using a map having the contents of FIG. 9 with A / N and N as parameters, whereas in this example, A / N is used. , N as independent parameters, the basic equilibrium amount M01AN and the rotation correction amount KN01 are obtained from the table containing the contents of FIG. 13 and FIG. 14, respectively, and their product (that is, KN01 corrects M01AN)
Is M01 (= M01AN × KN01).

Also in this example, as shown in FIG. 13, since A01 is used as a parameter, M01AN does not have such a sharply changing characteristic that the data quantity of the table is small as in the first embodiment.

In addition, two table values (M01AN, KN0
By using 1), although the accuracy is slightly inferior to the first embodiment in which a map having A / N and N as parameters is searched, the time required for the calculation of M01 can be greatly shortened. Responsiveness is improved.

(Effects of the Invention) As described above, according to the present invention, the equilibrium amount corresponding to the value obtained by dividing the opening area of the intake air by the engine speed and one other operating state signal is stored in advance as a map value. Every other time, the map value is retrieved from the division value at that time and another one of the operating state signals to obtain the equilibrium amount, or the basic equilibrium amount and other basic equilibrium amount corresponding to the value obtained by dividing the intake opening area by the engine speed The correction amount corresponding to one operating state signal is stored in advance as each table value, and the corresponding table value is searched from the division value at that time and another one operating state signal to correct the basic equilibrium amount and correction. Since the amount is calculated and the value obtained by correcting the basic equilibrium amount with the correction amount is used as the equilibrium amount, it is possible to avoid fluctuations in the equilibrium amount due to intake pulsation, and as a result, torque fluctuations can be reduced. Small vibration In addition, the amount of data is small, the resolution of grid points is also coarse, and the software and control unit capacity can be reduced.

[Brief description of drawings]

1 and 16 are block diagrams showing the conceptual configuration of the present invention. FIG. 2 is a mechanical block diagram of an embodiment of the present invention. 3 to 7 are flow charts showing the control contents of the air-fuel ratio control corresponding to the above embodiment. FIG. 8 is a characteristic diagram showing an example of the contents of a table that gives the opening area in the air-fuel ratio control, and FIG. 9 is the content of a map that gives the adhesion of the intake system and the equilibrium amount M0 of the floating fuel amount in the air-fuel ratio control. A characteristic diagram showing an example, FIG. 10 and FIG. 11 are characteristic diagram showing an example of the contents of a map that gives predetermined coefficients DKTw and DKN in the air-fuel ratio control. FIG. 12 is a flow chart illustrating a routine for calculating M01 according to the second embodiment of the present invention. FIG. 13 and FIG. 14 are characteristic diagrams showing the contents of the table giving the coefficients M01AN and KN01 in this embodiment. FIG. 15 is a characteristic diagram shown for comparison with the map characteristic diagram that gives M01 in the example of the present application. 1 ... Operating state detection means, 2 ... Basic injection amount calculation means,
3 ... Aperture area calculating means, 4 ... Division means, 5 ... Balance amount calculating means, 5A ... Balance amount storing means, 5B ... Balance amount reading means, 5C ... Basic balance amount storing means, 5D. Basic equilibrium amount reading means, 5E ... correction amount storage means, 5F ... correction amount reading means, 5G ... correction means, 6 ... difference value calculation means, 7 ... correction coefficient calculation means, 8 ... transient correction amount Calculation means, 9 ... Prediction variable calculation means, 10 ... Fuel quantity injection quantity calculation means, 11 ... Fuel supply means, 15 ... Throttle valve, 16 ... Intake passage, 17 ... Fuel injection valve, 20 ... Air flow rate sensor, 21 …… Crank angle sensor, 22 …… Water temperature sensor, 23 …… Air-fuel ratio sensor, 25 ……
Aperture sensor, 30 ... Control unit.

Claims (1)

(57) [Claims] 1. An operating state detecting means for detecting an operating state of an engine from parameters including at least an engine speed, an intake air amount and an engine temperature, and a basic fuel injection based on the engine speed and an intake air amount. A basic injection amount calculation means for calculating the amount, an opening area calculation means for calculating the opening area of the intake air from the intake throttle valve opening signal, a division means for dividing this opening area by the engine speed, and the opening area A value divided by the engine speed and an intake system adhesion corresponding to another one of the operating state signals, an equilibrium amount storage means for storing in advance the equilibrium amount of the floating fuel as a map value, and a division value obtained by the division means. Adhesion of intake system corresponding to the other one operation state signal, balance amount calculation means including balance amount reading means for reading balance amount of floating fuel from the balance amount storage means, or the opening area by engine speed. Adhesion of the intake system corresponding to the divided value, basic equilibrium amount storage means for pre-storing the basic equilibrium amount of floating fuel as a table value, basic equilibrium amount corresponding to the divided value obtained by the division means From the correction amount storage means, a correction amount storage means for storing in advance a correction amount for another one of the operating state signals as a table value, and a correction amount corresponding to another one of the operating state signals. A balance amount calculation unit including a correction amount reading unit and a correction unit that corrects the read basic equilibrium amount with the read correction amount as an equilibrium amount of adhering intake air and floating fuel; Difference value calculation means for calculating the difference between the adhered amount, the equilibrium amount of the floating fuel and the adherence of the intake system at that time, and the predictive variable of the floating fuel, and the difference value calculated by the difference value calculation means. A correction coefficient indicating how much is reflected in the correction of the amount, a correction coefficient calculation means for calculating based on the engine speed, the engine load and the engine temperature, and an excessive correction amount based on the difference value and the correction coefficient. Transient correction amount calculating means for calculating, the transient correction amount calculated by the transient correction amount calculating means and the adhesion,
Prediction variable calculation means for adding a prediction variable of floating fuel in synchronism with fuel injection and updating the prediction variable with the added value; basic injection amount calculated by the basic injection amount calculation means; and transient correction amount calculation means The fuel injection amount calculation means for calculating the fuel injection amount based on the transient correction amount calculated in step 1 and outputting the injection signal, and the fuel supply means for supplying the fuel to the engine based on the injection signal are provided. An air-fuel ratio control device for an internal combustion engine, which is characterized.