JP2007064230A - Cylinder inflow emission gas amount computing device of internal combustion engine and air intake pathway inflow emission gas computing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately compute an amount of an emission gas flowing into a cylinder of an internal combustion engine. <P>SOLUTION: An air intake pathway 13 and an emission gas pathway 19 are connected to each other via an emission gas recirculating pathway 21. Besides, an emission gas flow control valve 22 for controlling the amount of the emission gas flowing inside the emission gas recirculating pathway is arranged inside the emission gas recirculating pathway. A cylinder inflow emission gas amount that is the amount of the emission gas flowing into the cylinder 5 of the internal combustion engine is computed utilizing a passing emission gas amount that is the amount of the emission gas passing through the emission gas flow control valve. Further, the cylinder inflow emission gas amount is computed in consideration of a dead time corresponding to a time required for the emission gas that has passed through the emission gas flow control valve to reach the cylinder and of a following delay of a change in the cylinder inflow emission gas amount in relation to a change in the passing emission gas amount. This following delay is a primary delay and so a time constant and of dead time of the primary delay is allowed to change in response to an engine revolution speed. Therefore, the higher the engine revolution speed is, the smaller the primary delay and the dead are become. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an in-cylinder inflow exhaust gas amount calculation device and an intake passage inflow exhaust gas amount calculation device for an internal combustion engine.

An internal combustion engine in which an exhaust passage and an intake passage of an internal combustion engine are connected via an exhaust recirculation passage, and an exhaust gas flow rate control valve for controlling the flow rate of exhaust gas flowing in the exhaust recirculation passage is disposed in the exhaust recirculation passage However, this is disclosed in Patent Document 1. In this internal combustion engine, the exhaust gas discharged from the cylinder to the exhaust passage is introduced into the intake passage through the exhaust recirculation passage, whereby the exhaust gas can be recirculated into the cylinder.
In an internal combustion engine that recirculates exhaust gas in the cylinder in this way, for example, in order to determine the fuel injection amount so that the air-fuel ratio becomes the target air-fuel ratio, the exhaust gas is exhausted with respect to the gas amount filled in the cylinder. The ratio occupied by gas (so-called EGR rate) is calculated.

In Patent Document 1, the EGR rate during steady operation is obtained in the form of a map in advance, and the EGR rate is obtained from this map during steady operation. Moreover, in patent document 1, the EGR rate at the time of transient operation is estimated using the EGR rate obtained from the map of the EGR rate at the time of steady operation.
Specifically, when the amount of exhaust gas passing through the exhaust gas flow control valve is referred to as the passing gas amount, the exhaust gas flow control valve with respect to the passing gas amount obtained in relation to the target opening of the exhaust gas flow control valve. The EGR rate during transient operation is obtained by multiplying the EGR rate during steady operation by the ratio of the passing gas amount obtained in relation to the actual opening.

That is, the passage gas amount obtained in relation to the target opening of the exhaust gas flow control valve is represented by Qt, the passage gas amount obtained in association with the actual opening of the exhaust gas flow control valve is represented by Qa, and steady operation is performed. When the EGR rate at the time is represented by Rc, in Patent Document 1, the EGR rate R at the time of transient operation is obtained from R = Rc · Qa / Qt.
Further, in Patent Document 1, after an instruction to change the opening degree of the exhaust gas flow control valve is issued, there is some degree until the opening degree of the exhaust gas flow control valve actually reaches the target opening degree. The EGR rate at the time of transient operation calculated before this delay is used as the EGR rate at the time of transient operation.

JP-A-8-128359 JP 2002-130039 A JP 2001-41095 A

In Patent Document 1, the amount of exhaust gas that passes through the exhaust gas flow control valve is used to estimate the EGR rate during transient operation based on the EGR rate during steady operation, and the cylinder is filled. The amount of exhaust gas itself is not directly determined. For this reason, when the amount of exhaust gas charged in the cylinder is estimated from the EGR rate during the transient operation, it can be said that the accuracy of the estimated value is not so high.
Furthermore, there may be variations in the responsiveness of the exhaust gas flow control valve. In this case, the calculation accuracy of the gas amount itself passing through the exhaust gas flow control valve is lowered.
An object of the present invention is to accurately calculate the amount of exhaust gas flowing into a cylinder of an internal combustion engine.

In order to solve the above problems, in the first invention, an exhaust gas that connects an intake passage and an exhaust passage through an exhaust gas recirculation passage and controls the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage. In an internal combustion engine in which a flow rate control valve is disposed in the exhaust gas recirculation passage, an in-cylinder inflow exhaust gas amount that is an amount of exhaust gas flowing into the cylinder of the internal combustion engine is set to an exhaust gas that passes through the exhaust gas flow rate control valve. In the in-cylinder inflow exhaust gas amount calculation device that calculates the amount of passing exhaust gas that is the amount of gas, the dead time corresponding to the time it takes for the exhaust gas that has passed through the exhaust gas flow control valve to reach the cylinder In the in-cylinder inflow exhaust gas amount calculation device that calculates the in-cylinder inflow exhaust gas amount in consideration of the change in the in-cylinder inflow exhaust gas amount with respect to the change in the passing exhaust gas amount, A next delayed, the constant and the dead time when the first-order lag is changed in accordance with the engine rotational speed, the dead time is caused to decrease with decreasing high engine speed.
In order to solve the above problems, in a second aspect of the invention, an exhaust gas that connects an intake passage and an exhaust passage through an exhaust gas recirculation passage and controls the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage. In an internal combustion engine in which a flow control valve is disposed in an exhaust gas recirculation passage, a passing exhaust gas amount that is an amount of exhaust gas passing through the exhaust gas flow control valve is used using a parameter that changes the passing exhaust gas amount. In-cylinder inflow exhaust gas amount calculation device for calculating in-cylinder inflow exhaust gas amount, which is the amount of exhaust gas flowing into the cylinder of the internal combustion engine, using the calculated passing exhaust gas amount, The value of the parameter is read, and the read value corresponds to the dead time corresponding to the time taken for the exhaust gas that has passed through the exhaust gas flow control valve to reach the cylinder, and the change in the amount of exhaust gas that has passed. In the in-cylinder inflow exhaust gas amount calculation device that reflects the follow-up delay of the change in the in-flow exhaust gas amount and calculates the passing exhaust gas amount using the read value in which the dead time and the follow-up delay are reflected, The delay is a first-order delay, and the time constant of the first-order delay and the dead time are changed according to the engine speed, and the dead time is made smaller as the engine speed increases.
According to a third aspect, in the first or second aspect, the internal combustion engine includes a plurality of cylinders, and a ratio of an actual in-cylinder inflow exhaust gas amount in each cylinder to the calculated in-cylinder inflow exhaust gas amount. Is calculated in advance as a distribution coefficient, and the in-cylinder inflow exhaust gas amount in each cylinder is calculated by multiplying the calculated in-cylinder inflow exhaust gas amount by the distribution coefficient.
In a fourth aspect, in any one of the first to third aspects, the internal combustion engine includes a plurality of cylinders, and the follow-up delay and dead time are set for each cylinder.
According to a fifth aspect, in any one of the first to fourth aspects, the internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow control valve to the intake valve, or And means for changing the flow velocity of the gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage by changing the flow passage cross-sectional area of the intake passage controlled according to the engine operating state, and the passage volume Alternatively, the follow-up delay set value and the dead time set value are changed according to the gas flow rate.
In a sixth aspect of the invention, a function expression expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage is obtained and stored in advance, and the function expression is stored. In the first invention that uses the pressure in the intake passage to calculate the passing exhaust gas amount, the internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow control valve to the intake valve. Means, or means for changing the flow velocity of the gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage by changing the cross-sectional area of the intake passage according to the engine operating state, and the passage volume Alternatively, parameters other than the pressure in the intake passage are changed according to the gas flow rate.

In order to solve the above problems, in the seventh aspect of the invention, an exhaust gas that connects an intake passage and an exhaust passage through an exhaust gas recirculation passage and controls the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage. In an internal combustion engine in which a flow control valve is disposed in the exhaust gas recirculation passage, the amount of exhaust gas flowing into the intake passage, which is the amount of exhaust gas flowing into the intake passage, is determined by the amount of exhaust gas passing through the exhaust gas flow control valve. In the intake passage inflow exhaust gas amount calculation device that calculates the amount of exhaust gas passing through the exhaust gas, the dead time corresponding to the time it takes for the exhaust gas that has passed through the exhaust gas flow control valve to reach the intake passage And an intake passage inflow exhaust gas amount calculation device for calculating the intake passage inflow exhaust gas amount in consideration of a change in the intake passage inflow exhaust gas amount with respect to the change in the passing exhaust gas amount. In, the follow-up delay is a primary delay, the constant and the dead time when the first-order lag is changed in accordance with the engine rotational speed, the dead time is caused to decrease with decreasing high engine speed.
In order to solve the above problems, in an eighth aspect of the invention, an exhaust gas that connects an intake passage and an exhaust passage through an exhaust gas recirculation passage and controls the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage. In an internal combustion engine in which a flow control valve is disposed in an exhaust gas recirculation passage, a passing exhaust gas amount that is an amount of exhaust gas passing through the exhaust gas flow control valve is used using a parameter that changes the passing exhaust gas amount. In the intake passage inflow exhaust gas amount calculation device for calculating the inflow exhaust gas amount in the intake passage, which is the amount of exhaust gas flowing into the intake passage, using the calculated passing exhaust gas amount, The value of the parameter is read, and a change in the dead time corresponding to the time taken for the exhaust gas that has passed through the exhaust gas flow control valve to reach the intake passage and the amount of the exhaust gas that has passed is read into the read value. An intake passage inflow exhaust gas amount calculation device that reflects a follow-up delay of a change in the amount of exhaust gas inflow into the intake passage and calculates a passing exhaust gas amount using a read value in which the dead time and the follow-up delay are reflected , The follow-up delay is a first-order delay, and the time constant of the first-order delay and the dead time are changed according to the engine speed, and the dead time is reduced as the engine speed increases.
According to a ninth aspect, in the seventh or eighth aspect, the internal combustion engine includes a plurality of cylinders, and the exhaust gas recirculation passage is connected to an intake passage connected to each cylinder, and the above calculation is performed. The ratio of the actual intake passage inflow exhaust gas amount to the intake passage connected to each cylinder with respect to the intake passage inflow exhaust gas amount obtained in advance as a distribution coefficient is calculated in advance. The amount of exhaust gas flowing into the intake passage into the intake passage connected to each cylinder is calculated by multiplying the amount by the distribution coefficient.
According to a tenth aspect, in any one of the seventh to ninth aspects, the internal combustion engine includes a plurality of cylinders, and the follow-up delay and the dead time are set for each cylinder.
According to an eleventh aspect, in any one of the seventh to tenth aspects, the internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow control valve to the intake valve, or And means for changing the flow velocity of the gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage by changing the flow passage cross-sectional area of the intake passage controlled according to the engine operating state, and the passage volume Alternatively, the follow-up delay set value and the dead time set value are changed according to the gas flow rate.
In a twelfth aspect of the invention, a function equation expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage is obtained and stored in advance, and the function equation is stored. In the seventh aspect of the invention that uses the pressure in the intake passage to calculate the passing exhaust gas amount, the internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow control valve to the intake valve. Means, or means for changing the flow velocity of the gas in the vicinity of the opening of the exhaust gas recirculation passage opened to the intake passage by changing the flow passage cross-sectional area of the intake passage controlled according to the engine operating state, Parameters other than the pressure in the intake passage are changed according to the passage volume or the gas flow rate.

According to the first and second aspects of the invention, the cylinder inflow exhaust gas amount is directly calculated in consideration of the dead time and the follow-up delay. Therefore, the calculation accuracy of the cylinder inflow exhaust gas amount is high. Further, since the dead time and the time constant depend on the engine speed, the calculation accuracy of the in-cylinder inflow exhaust gas amount is higher.
According to the third aspect of the invention, even when the in-cylinder inflow exhaust gas amount varies among the cylinders, the calculation accuracy of the in-cylinder inflow exhaust gas amount is high for each cylinder.
According to the fourth aspect, even when there is variation in the dead time and follow-up delay for each cylinder, the calculation accuracy of the in-cylinder inflow exhaust gas amount is high for each cylinder.
If the passage volume from the exhaust gas flow control valve to the intake valve changes or the gas flow velocity near the opening of the exhaust gas recirculation passage that opens to the intake passage changes due to the change of the cross-sectional area of the intake passage, Although the dead time and the follow-up delay also vary, according to the fifth aspect, such factors are taken into account, so that the calculation accuracy of the in-cylinder inflow exhaust gas amount is high.
According to the sixth aspect, when the passing exhaust gas flow rate is calculated using a relational expression expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage. Therefore, parameters other than the pressure in the intake passage are changed according to the passage volume from the exhaust gas flow control valve to the intake valve or the gas flow velocity in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage. The calculation accuracy of the passing exhaust gas amount calculated from the above relational expression is high.

According to the seventh and eighth aspects, since the exhaust gas amount flowing into the intake passage is directly calculated in consideration of the dead time and the follow-up delay, the calculation accuracy of the exhaust gas amount flowing into the intake passage is high. It can be said. Further, since the dead time and the time constant depend on the engine speed, the calculation accuracy of the inflow exhaust gas amount in the intake passage is higher.
According to the ninth aspect, even when there is a variation in the inflow exhaust gas amount in the intake passage for each cylinder, the calculation accuracy of the inflow exhaust gas amount in the intake passage is high for each cylinder.
According to the tenth aspect, even when there is variation in the dead time and follow-up delay for each cylinder, the calculation accuracy of the inflow exhaust gas amount in the intake passage is high for each cylinder.
If the passage volume from the exhaust gas flow control valve to the intake valve changes, or if the gas flow velocity near the opening of the exhaust gas recirculation passage that opens to the intake passage changes due to the change in the cross-sectional area of the intake passage, Although the dead time and follow-up delay also vary, according to the eleventh aspect, such factors are taken into account, so that the calculation accuracy of the amount of exhaust gas flowing into the intake passage is high.
According to the twelfth invention, when the passing exhaust gas flow rate is calculated using a relational expression expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage. Therefore, parameters other than the pressure in the intake passage are changed according to the passage volume from the exhaust gas flow control valve to the intake valve or the gas flow velocity in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage. The calculation accuracy of the passing exhaust gas amount calculated from the above relational expression is high.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is an overall view of an internal combustion engine to which the present invention is applied. In the following description, a cylinder injection type spark ignition type internal combustion engine will be described as an example. However, the present invention can be applied to other spark ignition type internal combustion engines and compression self-ignition type internal combustion engines.
In FIG. 1, 1 is an engine body, 2 is a cylinder block, 3 is a piston, 4 is a cylinder head, 5 is a cylinder (combustion chamber), 6 is an intake valve, 7 is an intake port, 8 is an exhaust valve, and 9 is an exhaust port. Reference numeral 10 denotes a spark plug, 11 denotes a fuel injection valve, and 12 denotes a cavity.

The intake port 7 is connected to the surge tank 14 via the intake branch pipe 13 for each cylinder 5. The surge tank 14 is connected to an air cleaner 16 via an upstream side intake pipe 15. A throttle valve 18 driven by a step motor 17 is disposed in the upstream side intake pipe 15.
On the other hand, the exhaust port 9 is connected to an exhaust pipe 19. The exhaust pipe 19 is connected to the exhaust purification device 20.

  The exhaust pipe 19 is connected to the intake branch pipe 13 via an exhaust gas recirculation passage (hereinafter referred to as an EGR passage). Further, an EGR control valve 22 that controls the flow rate of exhaust gas flowing through the EGR passage 21 is disposed in the EGR passage 21. Exhaust gas discharged from the cylinder 5 of the internal combustion engine is again introduced into the cylinder through the EGR passage 21. In the present specification, the exhaust gas recirculated into the cylinder through the EGR passage 21 is also referred to as EGR gas.

An electronic control unit (ECU) 31 comprises a digital computer, and is connected to each other via a bidirectional bus 32, a RAM (random access memory) 33, a ROM (read only memory) 34, a CPU (microprocessor) 35, an input. A port 36 and an output port 37 are provided.
An intake pipe pressure sensor 40 for detecting the pressure in the intake branch pipe 13 (hereinafter referred to as intake pipe pressure) is attached to the intake branch pipe 13. The intake pipe pressure sensor 40 generates an output voltage proportional to the intake pipe pressure, and this output voltage is input to the input port 36 via the corresponding AD converter 38.

  The internal combustion engine also has a throttle opening sensor 43 for detecting the opening of the throttle valve 18, the atmospheric pressure around the internal combustion engine, or the pressure of the air taken into the upstream intake pipe 15 (intake pressure). And an atmospheric temperature sensor 45 for detecting the temperature of the atmosphere around the internal combustion engine or the temperature of the air taken into the upstream side intake pipe 15 (intake air temperature). . These sensors 44 and 45 generate output voltages proportional to the atmospheric pressure and the atmospheric temperature, respectively, and these output voltages are input to the input port 36 via the corresponding AD converters 38.

A load sensor 47 that generates an output voltage proportional to the amount of depression of the accelerator pedal 46 is connected to the accelerator pedal 46. The output voltage of the load sensor 47 is input to the input port 36 via the corresponding AD converter 38.
The internal combustion engine includes a crank angle sensor 48. For example, the crank angle sensor 48 generates an output pulse every time the crankshaft rotates 30 degrees, and the output pulse is input to the input port 36. The CPU 35 calculates the engine speed from the output pulse of the crank angle sensor 45.

  On the other hand, the output port 37 is connected to the spark plug 10, the fuel injection valve 11, the step motor 17, and the EGR control valve 22 via a corresponding drive circuit 39. The opening degree of the EGR control valve 22 is calculated in the ECU 31 based on a step signal sent from the output port 37 to the EGR control valve 22.

By the way, the amount of fuel to be injected into the cylinder 5 from the fuel injection valve 10 (hereinafter simply referred to as fuel injection amount) is based on the amount of air charged in the cylinder 5 and the mixed gas in the cylinder 5. The air-fuel ratio is determined to be the target air-fuel ratio. Therefore, in order to accurately set the air-fuel ratio of the mixed gas in the cylinder 5 to the target air-fuel ratio, it is necessary to accurately grasp the amount of air charged in the cylinder 5 (hereinafter referred to as in-cylinder charged air amount). There is.
Here, as a method for estimating the amount of air filled in the cylinder, an air flow meter for detecting the mass flow rate of air passing through the throttle valve 18 and other sensors are attached to the internal combustion engine, and the output values of these sensors are used as variables. There is a method of preparing a map for calculating the in-cylinder charged air amount in advance and estimating the in-cylinder charged air amount from the output value of the sensor and the map.

However, when estimating the in-cylinder charged air amount using a map, in order to accurately estimate the in-cylinder charged air amount, the number of maps and their arguments must be increased. However, if the number of maps is increased, the storage capacity of the ROM 34 for storing these maps must be increased, which increases the cost of the internal combustion engine.
Furthermore, when estimating the cylinder charge air amount using a map, it is necessary to create a map for each internal combustion engine model, or for each internal combustion engine even in the same model internal combustion engine. The map creation effort increases.
On the other hand, if the number of map arguments is increased in order to more accurately estimate the amount of air charged in the cylinder, the map creation labor is greatly increased.

Therefore, a method for calculating the in-cylinder charged air amount by numerical calculation using an expression derived from a model instead of a map has been studied, and such a method has already been filed by the present applicant (Japanese Patent Application No. 2001-316350). issue).
Since the method already filed by the applicant of the present application is a method applied to an internal combustion engine in which exhaust gas is not recirculated in the cylinder, it cannot be directly applied to the internal combustion engine with the EGR device of this embodiment. However, since it will be helpful in understanding a method to be described later that can be applied to the present embodiment, a method already filed by the applicant of the present application will be described.

The method already filed by the applicant of the present application is derived from the model shown in FIG. 2 (hereinafter referred to as a cylinder inflow gas amount model).
The cylinder inflow gas amount model M20 shown in FIG. 2 includes a throttle model M21, an intake pipe model M22, and an intake valve model M23.

The throttle model M21, the opening degree (throttle opening) theta _t of the throttle valve 18 detected by a throttle opening sensor 43, the atmospheric pressure P _a, which is detected by the atmospheric pressure sensor 45, detected by the atmospheric temperature sensor 44 and the atmospheric temperature T _a, which is the pressure in the intake tube 13 calculated in the intake pipe model M22 (hereinafter, the intake pipe referred to as pressure) and P _m are inputted, in the model M21, from these input parameters , the flow rate of air passing through the throttle valve 18 per unit time (hereinafter referred to as the throttle valve passing air flow rate) m _t is calculated.

The intake pipe model M22, the throttle valve passage air flow rate m _t calculated at the throttle model M21 as described above, the flow rate of the gas flowing into the cylinder 5 to the per unit time calculated at the intake valve model M23 (hereinafter , it referred to as cylinder inflow gas flow rate) and m _c, are input and the atmospheric temperature T _a, in the model M22, from these input parameters, the intake pipe pressure P _m, the temperature of the gas in the intake branch pipe 13 _Tm (hereinafter referred to as intake pipe temperature) is calculated.

The intake valve model M23, the intake pipe pressure P _m calculated at the intake pipe model M22 described above, the intake pipe temperature T _m, is input and the atmospheric temperature T _a, in the model M23, are those input The in-cylinder inflow gas flow rate _mc is calculated from the obtained parameters.
Then, in this method, as described later, the amount of gas flowing by utilizing the cylinder inflow gas flow rate m _c into the cylinder 5 (hereinafter, referred to as in-cylinder charged gas amount) M _c is calculated.

As can be seen from FIG. 2, in the in-cylinder inflow gas amount model M20, the parameter value calculated in each model is used as a parameter value input to another model. parameter value is input, the throttle opening theta _t, the atmospheric pressure P _a, and is only three parameters of the atmospheric temperature T _a. That is, according to this method, it can be said that the in-cylinder charged gas amount _Mc is calculated from the three parameters.

式（１）において、μ_tはスロットル弁における流量係数であり、スロットル開度θ_tの関数であって、図３に示したマップから定まる。また、Ａ_tはスロットル弁の開口断面積であり、スロットル開度θ_tの関数であって、図４に示したマップから定まる。なお、これら流量係数μ_tおよび開口断面積Ａ_tをまとめたμ_t・Ａ_tをスロットル開度θ_tの関数で１つのマップから求めるようにしてもよい。また、Ｒは気体定数に関する定数であり、いわゆる気体定数Ｒ^*を１モル当たりの空気の質量Ｍ_aで除算した値である（Ｒ＝Ｒ^*／Ｍ_a）。 Next, each of the models M21 to M23 will be described in detail.
In the throttle model M21, enter the atmospheric pressure P _a and the atmospheric temperature T _a and the intake pipe pressure P _m and the throttle opening theta _t in equation (1), by solving this equation, the throttle valve passage air flow rate m _t is calculated.

In the equation (1), μ _t is a flow coefficient in the throttle valve, is a function of the throttle opening θ _t , and is determined from the map shown in FIG. Also, A _t is the cross-sectional area of the opening of the throttle valve, a function of the throttle opening theta _t, determined from the map shown in FIG. It may be determined from a single map these flow coefficient mu _t and the opening cross-sectional area A _t summarizes mu _t · A _t as a function of the throttle opening theta _t. Further, R is a constant related to a gas constant is a value obtained by dividing the mass M _a of air per 1 mole of the so-called gas constant ^{R * (R = R * /} M a).

式（２）において、κは比熱比であり、この方法では、一定値としている。 Further, Φ (P _m / P _a ) is a function having P _m / P _a as a variable, as shown in the following equation (2).

In the equation (2), κ is a specific heat ratio, and is a constant value in this method.

Note that there is a relationship as shown in FIG. 5 between the function Φ (P _m / P _a ) and P _m / P _a . Therefore, instead of the equation (2), a map for calculating _a function Φ (P _m / P _a ) having P _m / P _a as a variable is stored in the ROM 34 in advance, and P _m / P _a and this map From the above, the value of the function Φ (P _m / P _a ) may be calculated.

Note that these equations (1) and (2), the air pressure of the throttle valve 18 upstream of the air and the atmospheric pressure P _a, the temperature of the throttle valve 18 upstream of the air and the atmospheric temperature T _a, passes through the throttle valve 18 6 is the intake pipe pressure P _m , and the throttle valve 18 is based on the model as shown in FIG. 6, the law of conservation of mass, energy between the air upstream of the throttle valve 18 and the air that has passed through the throttle valve 18. Using the conservation equation and the relational expression established in the momentum conservation law, the equation of state of gas, the specific expression of the specific heat ratio (κ = Cp / Cv), and the Meyer's relational expression (Cp = Cv + R ^* ) Derived. Here, Cp is a constant pressure specific heat, Cv is a quantitative specific heat, and R ^* is a so-called gas constant.

式（３）および（４）において、Ｖはスロットル弁１８から吸気弁６までの吸気管１５、サージタンク１４、吸気枝管１３、および、吸気ポート７（以下、これらまとめて、吸気管部分と称す）のトータルの容積であり、通常、一定値である。 Next, the intake pipe model M22 will be described. In the intake pipe model M22, by inputs the throttle valve passage air flow rate m _t and the cylinder inflow gas flow rate m _c and the atmospheric temperature T _a to the following equation (3) and (4), solving these equations, the intake pipe The pressure P _m and the intake pipe temperature T _m are calculated.

In equations (3) and (4), V is the intake pipe 15 from the throttle valve 18 to the intake valve 6, the surge tank 14, the intake branch pipe 13, and the intake port 7 (hereinafter collectively referred to as the intake pipe portion and Generally), and is a constant value.

  These equations (3) and (4) are related to the intake pipe portion, and the air flowing into the intake pipe portion and the air flowing out from the intake pipe portion and flowing into the cylinder are based on the model shown in FIG. It is derived from the relational expression that holds in the law of conservation of mass and the law of conservation of energy.

そして、この式（５）と、気体の状態方程式（Ｐ_m・Ｖ＝Ｍ・Ｒ^*・Ｔ_m）とから、上記式（３）が導き出される。 Specifically, if the total air amount in the intake pipe portion is M, the temporal change in the total air amount M is the flow rate of air flowing into the intake pipe portion (that is, the throttle valve passage air flow rate) m _t. and the flow rate of air flowing into the outflow to the cylinder from the intake pipe portion (i.e., the cylinder inflow gas flow rate) to be equal to the difference between m _c, the mass conservation law, the following equation (5) is satisfied.

Then, the equation (3) is derived from the equation (5) and the gas equation of state ( _Pm · V = M · R ^* · T _m ).

そして、この式（６）と、上述した気体の状態方程式とから、上記式（４）が導き出される。 Further, the temporal variation amount of energy M · Cv · T _m of the air in the intake pipe portion, the energy of the air entering the outlet to the cylinder from the energy amount and the intake pipe of the air flowing into the intake pipe portion since equal to the difference between the amount and the temperature of the air flowing into the intake pipe portion and the atmospheric temperature T _a, the temperature of the air flowing into the cylinder flows out from the intake pipe portion and the intake pipe temperature T _m, the energy The following equation (6) is established on the conservation law.

Then, the equation (4) is derived from the equation (6) and the above-described gas state equation.

式（７）において、ａおよびｂは、機関回転数ＮＥを変数として求まる値である。また、内燃機関において、吸気弁６の開弁タイミングまたは閉弁タイミングに相当するバルブタイミング、および、開弁時間に相当する作用角の少なくとも一方が変更可能となっている場合には、式（７）において、ａおよびｂは、機関回転数ＮＥとバルブタイミングまたは位相角またはこれら両方とを変数として求まる値である。
別の云い方をすれば、式（７）において、ａは比例係数であり、ｂは排気弁８の閉弁時に気筒５内に残存していたガスの量を表す値である。 Next, the intake valve model M23 will be described. In the intake valve model M23, the intake pipe pressure P _m and the intake pipe temperature T _m and the atmospheric temperature T _a is input to the following equation (7), by solving this equation, the cylinder inflow gas flow rate m _c is calculated The

In Expression (7), a and b are values obtained using the engine speed NE as a variable. In the internal combustion engine, when at least one of the valve timing corresponding to the valve opening timing or the valve closing timing of the intake valve 6 and the operating angle corresponding to the valve opening time can be changed, the equation (7) ), A and b are values obtained by using the engine speed NE and the valve timing and / or the phase angle as variables.
In other words, in Equation (7), a is a proportional coefficient, and b is a value representing the amount of gas remaining in the cylinder 5 when the exhaust valve 8 is closed.

In the formula (7), when the engine operating condition has changed, i.e., at the time of transient operation, since the intake pipe temperature T _m may change significantly, to compensate for changes in the intake pipe temperature T _m as a correction coefficient for, it is used T _a / T _m derived from theory and empirical rules.

Note that equation (7) relates to the intake valve 6, and the basis of the model as shown in FIG. 8, to be proportional to the cylinder inflow gas flow rate m _c to the intake pipe pressure P _m as described in detail below Deemed and derived from theory and empirical rules.
That is, the in-cylinder charged gas amount M _c is proportional to the pressure in the cylinder 5 at the time of closing of the valve closing finalized and the intake valve during 6 of the intake valve 6. Here, since the pressure in the cylinder 5 when the intake valve 6 is closed can be regarded as equal to the pressure of the air upstream of the intake valve 6 (that is, the intake pipe pressure) P _m , the in-cylinder charged gas amount _Mc is the intake air. It can be approximated to be proportional to the tube pressure P _m .

On the other hand, the in-cylinder charged gas amount M _c is the air flow rate (cylinder inflow gas flow rate) m _c flowing into the cylinder 5 during the valve opening period of the intake valve 6 over the valve opening period of the intake valve 6. It is obtained by time integration. Namely, relationship that is between the in-cylinder charged gas amount M _c and the cylinder inflow gas flow rate m _c, the time integral value of the cylinder inflow gas flow rate m _c is cylinder charged gas amount M _c.
Thus, in-cylinder charged gas amount M _c is proportional to the intake pipe pressure P _m, and, between the in-cylinder charged gas amount M _c and the cylinder inflow gas flow rate m _c, the cylinder inflow gas flow rate m _c It can be regarded as the time integral value of from that there is a relationship that a cylinder charged gas amount M _c, also inflow gas flow rate m _c-cylinder proportional to the intake pipe pressure P _m.

Therefore, in this method, it is assumed that the in-cylinder inflow gas flow rate _mc is proportional to the intake pipe pressure P _m , and Equation (7) is derived from theory and empirical rules.
Note that the cylinder inflow gas flow rate m _c calculated by Equation (7) are the average value of the flow rate of air flowing out from the intake pipe portion per unit time, the cylinder inflow gas flow rate m _c, the internal combustion engine by applying a time divided by the number of cylinders of the time required for one cycle, the in-cylinder charged gas amount M _c in each cylinder 5 is calculated.

Next, this will be described with reference to FIG. 9, taking an internal combustion engine having four cylinders as an example.
In FIG. 9, the horizontal axis is the crank angle, and the vertical axis is the amount of air flowing from the intake pipe portion into the cylinder 5 per unit time. In the example shown in FIG. 9, the intake stroke is performed in the order of the first cylinder # 1, the third cylinder # 3, the fourth cylinder # 4, and the second cylinder # 2. When the intake stroke is performed in this way, the flow rate of air flowing into each cylinder 5 from the intake pipe portion changes as indicated by the broken line in FIG. 9, and as a result, the flow rate of air flowing out from the intake pipe portion is 9 will change as indicated by the solid line in FIG.

The average value of the flow rate of air flowing out from the intake pipe portion (solid line in FIG. 9) is the in-cylinder inflow gas flow rate _mc , and is indicated by a broken line in FIG. Therefore, the cylinder charge gas amount M _c in each cylinder 5 is equal to the cylinder inflow gas flow rate m _c (broken line in FIG. 9) and the time taken for one cycle of the internal combustion engine (in the example shown in FIG. Time taken to rotate 720 °) divided by the number of cylinders (four in the example shown in FIG. 9), that is, time taken for the crankshaft to rotate 180 ° in the example shown in FIG. It is calculated by applying. Then, the cylinder charged gas amount M _c in each cylinder 5 calculated by Thus, for example, will correspond to the hatched in FIG.

Next, a method of calculating the cylinder charged gas amount M _c when the cylinder inflow gas amount model M20 described above is mounted on an internal combustion engine.
The in-cylinder charged gas amount _Mc is obtained from the equations (1) to (4) and (7) of each model of the in-cylinder inflow gas amount model M20. These five equations are used when mounted on an internal combustion engine. , It is discretized so that it can be processed by the ECU 31. That is, assuming that the time is _t and the calculation interval (calculation period) is Δt, these five expressions are discretized into the following expressions (8) to (12).

According to the in-cylinder inflow gas amount model M20 discretized and mounted in the internal combustion engine, the throttle valve passing air flow rate m _t (t) at time t calculated in the throttle model M21 and the intake valve model M23 The in-cylinder inflow gas flow rate m _c (t) at time t and the intake pipe temperature T _m (t) calculated at time t are input to the equations (10) and (11) of the intake pipe model M22. By solving Expressions (10) and (11), the intake pipe pressure P _m (t + Δt) and the intake pipe temperature T _m (t + Δt) at time (t + Δt) are calculated.

Then, the intake pipe pressure P _m (t + Δt) calculated in the intake pipe model M22 and the throttle opening θ _t (t) at the time t are input to the equations (8) and (9) of the throttle model M21. Is calculated, the throttle valve passing air flow rate m _t (t + Δt) at time (t + Δt) is calculated.
Further, the intake pipe pressure P _m (t + Δt) and the intake pipe temperature T _m (t + Δt) calculated in the intake pipe model M22 are input to the expression (12) of the intake valve model M23, and the time ( The cylinder inflow gas flow rate m _c (t + Δt) at t + Δt) is calculated.

By repeating such calculations, the cylinder inflow gas flow rate m _c at any time is calculated. Then, the in-cylinder inflow gas flow rate m _c thus calculated is multiplied by the time obtained by dividing the time taken for one cycle by the number of cylinders, as described above, so that the in-cylinder charged gas amount M of each cylinder at an arbitrary time is obtained. _c is calculated.

Incidentally, when the internal combustion engine is started, i.e., at time t = 0, the intake pipe pressure P _m is equal to the atmospheric pressure _{P a (P m (0)} = P a), whereas, the intake pipe temperature T _m It is set equal to the atmospheric temperature T _a (T _m (0) = T _a ), and calculation in each of the models M21 to M23 is started.
Further, as the atmospheric pressure P _a and the atmospheric temperature T _a is used in the cylinder inflow gas amount model M20 as described above, to always be using an atmospheric pressure and the atmospheric temperature when the calculation of the model M20 is started, Atmospheric pressure P _a (t) and atmospheric temperature T _a (t) at time _t may be used.

式（１３）において、ｍ_cは気筒内に流入するガスの流量である筒内流入ガス流量であり、ｍ_c-egrは気筒内に流入するガス中の排気ガスの流量である筒内流入ＥＧＲガス流量である。 Next, a cylinder inflow gas amount model applicable to the internal combustion engine with the EGR device shown in FIG. 1 will be described.
When referred to the flow rate of the cylinder inflow fresh air flow rate m _c-air air in the gas flowing into the cylinder (fresh air), the cylinder inflow fresh air flow rate m _c-air is found from the following equation (13).

In Expression (13), m _c is a cylinder inflow gas flow rate that is a flow rate of gas flowing into the cylinder, and m _c-egr is a cylinder inflow EGR that is a flow rate of exhaust gas in the gas flowing into the cylinder. The gas flow rate.

Here, when the engine is in steady operation (for example, when the throttle opening, EGR opening, engine speed, etc. are maintained substantially constant), the flow rate of exhaust gas passing through the EGR control valve 22 (hereinafter referred to as EGR). _{M egr (} referred to as control valve passage gas flow rate) and in-cylinder inflow EGR gas flow rate m _c-egr are equal ( _mc-egr = m _egr ).
For this reason, Formula (13) can be rewritten to following Formula (14).

Therefore, Motomare and the cylinder inflow gas flow rate m _c and the EGR control valve passage gas flow rate m _egr, Motomari is the cylinder inflow fresh air flow rate m _c-air, therefore, the cylinder 5 at the time of closing of the intake valves 6 An in-cylinder charged fresh air amount _Mc-air, which is the amount of air (fresh air) filled, is obtained.
Therefore, in the first embodiment, as the cylinder inflow gas amount model for calculating the cylinder inflow gas flow rate m _c, using a model M10 shown in FIG. 10.
The cylinder inflow gas amount model M10 shown in FIG. 10 includes a throttle model M11, an intake pipe model M12, an intake valve model M13, and an EGR control valve model M14.

The throttle model M11 is the same model as the throttle model M21 shown in FIG. 2, and the intake valve model M13 is also the same model as the intake valve model M23 shown in FIG. to describe in, in the throttle model M11, and inputs the throttle opening theta _t and the atmospheric pressure P _a and the atmospheric temperature T _a and the intake pipe pressure P _m in the formula (1), by solving this equation, it passes the throttle valve calculated air flow rate m _t is the intake valve model M13, inputs the atmospheric temperature T _a and the intake pipe pressure P _m and the intake pipe temperature T _m in the formula (7), by solving this equation, the cylinder inflow A gas flow rate _mc is calculated.

The intake pipe model M12, the throttle valve passage air flow rate m _t calculated at the throttle model M11, the cylinder inflow gas flow rate m _c calculated at the intake valve model M13, the atmospheric temperature T _a, EGR control valve model The flow rate of exhaust gas passing through the EGR control valve 22 per unit time calculated in M14 (hereinafter referred to as EGR control valve passage gas flow rate) m _egr is input, and in the model M12, from these input parameters The intake pipe pressure P _m and the intake pipe temperature T _m are calculated.

Further, the EGR control valve model M14 is an EGR opening degree theta _e and the atmospheric pressure P _a and the atmospheric temperature T _a and the intake pipe pressure P _m calculated at the intake pipe model M12 and the exhaust temperature T _e is input, the In the model M15, the EGR control valve passage gas flow rate m _egr is calculated from these input parameters.
As can be seen from FIG. 10, in the in-cylinder inflow gas amount model M10, the parameter value calculated in each model is used as a parameter value input to another model. parameter value is input, the throttle opening theta _t, EGR opening degree theta _e, the atmospheric pressure P _a, and is only four parameters of the atmospheric temperature T _a. That is, according to this embodiment, it can be said that the cylinder charge gas amount _Mc is calculated from these four parameters.

式（１５）および（１６）においても、Ｖはスロットル弁１８から吸気弁６までの吸気管部分の容積であり、通常、一定値である。 Next, the intake pipe model M12 and the EGR control valve model M14 will be described in detail.
In the intake pipe model M12, the throttle valve passage air flow rate m _t and the cylinder inflow gas flow rate m _c and the atmospheric temperature T _a and the EGR control valve passage gas flow rate m _egr by the following equation and an exhaust temperature T _e (15) and (16 ) And solving these equations, the intake pipe pressure P _m and the intake pipe temperature T _m are calculated.

Also in the equations (15) and (16), V is the volume of the intake pipe portion from the throttle valve 18 to the intake valve 6 and is usually a constant value.

  As described above with respect to the intake pipe model M22, these equations (15) and (16) relate to the intake pipe portion and flow into the intake pipe portion based on the model shown in FIG. It is derived from a relational expression that is established based on the law of conservation of mass and the law of conservation of energy among the air that flows, the exhaust gas that flows into the intake pipe portion, and the gas that flows out of the intake pipe portion and flows into the cylinder.

式（１７）において、μ_eはＥＧＲ制御弁２２における流量係数であって、ＥＧＲ開度θ_eの関数である。また、Ａ_eはＥＧＲ制御弁２２の開口断面積であって、ＥＧＲ開度θ_eの関数である。また、Ｒは気体定数に関する定数であり、いわゆる気体定数Ｒ^*を１モル当たりの排気ガスの質量Ｍ_eで除算した値である（Ｒ_e＝Ｒ^*／Ｍ_e）。 Further, the EGR control valve model M14, by inputting the EGR opening theta _e and the intake pipe pressure P _m and the exhaust pressure P _e and the exhaust temperature T _e in the following equation (17), the EGR control valve passage gas flow rate m _egr is calculated.

In the equation (17), μ _e is a flow coefficient in the EGR control valve 22 and is a function of the EGR opening degree θ _e . A _e is the opening cross-sectional area of the EGR control valve 22 and is a function of the EGR opening degree θ _e . Further, R is a constant related to a gas constant is a value obtained by dividing the mass M _e of the exhaust gas per 1 mol of a so-called gas constant ^{_{R * (R e = R *}} / M e).

この式（１８）は、式（２）の変数Ｐ_m／Ｐ_aを変数Ｐ_m／Ｐ_eに置き換えたものである。したがって、κは比熱比であり、本実施形態では、一定値である。 _Pe is the pressure of the exhaust gas upstream of the EGR control valve 22, and _Te is the temperature of the exhaust gas upstream of the EGR control valve 22.
Further, Φ (P _m / P _e ) is a function having P _m / P _e as a variable, as shown in the following equation (18).

This equation (18) is obtained by replacing the variable P _m / P _a in the equation (2) with the variable P _m / P _e . Therefore, κ is a specific heat ratio, and is a constant value in this embodiment.

Also, there is a relationship as shown in FIG. 5 between the function Φ (P _m / P _e ) and P _m / P _e . Therefore, instead of equation (18), stored in advance in the function Φ (P _m / P _e) map for calculating the ROM34 to variable P _m / P _e, and the map and P _m / P _e From the above, the value of the function Φ (P _m / P _e ) may be calculated.

Note that these equations (17) and (18), the pressure of the EGR control valve 22 upstream of the exhaust gas and the exhaust pressure P _e, the temperature of the EGR control valve 22 upstream of the exhaust gas and the exhaust temperature T _e, the EGR control valve as an intake pipe pressure P _m and the pressure of the exhaust gas passing through the 22 relates EGR control valve, in the basic model shown in FIG. 12, has passed through the EGR control valve 22 upstream of the exhaust gas and the EGR control valve 22 It is derived by using the relational expression that is established with respect to the exhaust gas in terms of the law of conservation of mass, the law of conservation of energy, and the law of conservation of momentum, as well as the equation of state of gas, the definition of specific heat ratio, and the Meyer's relational expression. .
That is, the basic idea for deriving these equations (17) and (18) is the same as the idea for deriving equations (1) and (2) for calculating the throttle valve passage air flow rate. It is.

Thus, according to this embodiment, by the high pressure P _e and the exhaust temperature T _e and the intake pipe pressure P _m and the EGR opening theta _e entered in the formula and numerical calculation, EGR control valve passage Since the gas flow rate is calculated and therefore the number of maps used at this time is reduced, the map creation effort can be greatly reduced.

Next, a method for calculating the in _- cylinder charged fresh air amount _Mc-air when the in-cylinder inflow gas amount model M10 is mounted on the internal combustion engine will be described.
The in-cylinder charged fresh air amount _Mc-air is obtained from the equations (1), (2), (7), and (15) to (18) of each model of the in-cylinder inflow gas amount model M10. When being mounted on an internal combustion engine, it is discretized so that it can be processed by the ECU 31. That is, when time is _t and calculation interval (calculation cycle) is Δt, equations (1), (2), and (7) are discretized into equations (8), (9), and (12), respectively. .
Moreover, Formula (15)-(18) is discretized by following Formula (19)-(22), respectively.

Further, equations for calculating the exhaust pressure _Pe and the exhaust temperature _Te are also discretized as the following equations (23) and (24). Details regarding these equations will be described later.

According to the in-cylinder inflow gas amount model M10 discretized and implemented in the internal combustion engine, the throttle valve passing air flow rate m _t (t) at the time t calculated in the throttle model M11 and the EGR control valve model EGR control valve passage gas amount m _egr (t) at time t calculated in M14, in-cylinder inflow gas flow rate m _c (t) at time t calculated in intake valve model M13, and exhaust temperature T at time _{t e} (t) is input to the equations (19) and (20) of the intake pipe model M12, and by solving these expressions, the intake pipe pressure P _m (t + Δt) and the intake pipe temperature T _m (at time (t + Δt)) t + Δt) is calculated.

Then, the intake pipe pressure P _m (t + Δt) at the time (t + Δt) calculated in the intake pipe model M12 and the throttle opening θ _t (t + Δt) at the same time (t + Δt) are expressed by the equation (8) of the throttle model M11 and By inputting into (9) and solving these equations, the throttle valve passing air flow rate m _t (t + Δt) at time (t + Δt) is calculated.
Further, the intake pipe pressure P _m (t + Δt) and the intake pipe temperature T _m (t + Δt) at the time (t + Δt) calculated in the intake pipe model M12 are input to the expression (12) of the intake valve model M13, and this expression is solved. Thus, the in-cylinder inflow gas flow rate m _c (t + Δt) at the time (t + Δt) is calculated.

Then, the in-cylinder inflow gas flow rate m _c (t + Δt) at the time (t + Δt) calculated in the intake pipe model M12 and the EGR control valve passage gas flow rate m at the time t calculated in the EGR control valve model M14 in the previous routine. _By entering _egr (t) into equation (14) and solving this equation, the cylinder inflow fresh air flow rate _mc-air (t + Δt) at time (t + Δt) is calculated.

Then, the intake pipe pressure P _m (t + Δt) at the time (t + Δt) calculated in the intake pipe model M12, the exhaust pressure _Pe (t + Δt) at the same time (t + Δt), and the EGR opening θ at the same time (t + Δt). _e (t + Δt) is input to the equations (21) and (22) of the EGR control valve model M14, and these equations are solved to calculate the EGR control valve passage gas flow rate m _egr (t + Δt) at time (t + Δt). The
Further, the engine load factor KL (t + Δt) at time (t + Δt) and the engine speed NE at the same time (t + Δt) are input to equation (23), and the exhaust pressure at time (t + Δt) is solved by solving this equation. P _e (t + Δt) is calculated.

Further, the EGR control valve passage gas flow rate m _egr (t) at time t and the engine speed NE (t + Δt) at time (t + Δt) are input to equation (24), and by solving this equation, time (t + Δt) ) At the exhaust temperature T _e (t + Δt).
The calculated m _c (t + Δt), m _egr (t + Δt), and m _t (t + Δt) are again input to the equations (19) and (20) of the intake pipe model M12.
By repeating such calculation, the in-cylinder fresh air flow rate _mc-air at the time is arbitrarily calculated. Then, as described above, the in _- cylinder fresh air flow rate _mc-air calculated in this way is multiplied by the number of cylinders divided by the time required for one cycle, so that the in-cylinder filling of each cylinder at an arbitrary time is performed. A fresh air amount Mc _-air is calculated.

Incidentally, when the internal combustion engine is started, i.e., at time t = 0, the intake pipe pressure P _m is equal to the atmospheric pressure _{P a (P m (0)} = P a), the intake pipe temperature T _m and exhaust temperature T _e is equal to the atmospheric temperature _{T a (T m (0)} = T a, T e (0) = T a), is calculated for each model M11~M13 is started.
Further, as the atmospheric pressure P _a and the atmospheric temperature T _a is used in the cylinder inflow gas amount model M10 as described above, to always be using an atmospheric pressure and the atmospheric temperature when the calculation of the model M10 is started, Atmospheric pressure P _a (t) and atmospheric temperature T _a (t) at time t may be used.

Further, in the present embodiment, the exhaust pressure P _e, the exhaust temperature T _e, the intake pipe pressure P _m, and, from equation (17) utilizing the four parameters of the EGR opening degree theta _e, EGR control valve passage gas flow rate m _egr is calculated, but these four parameters are used, but the EGR control valve passage gas flow rate m _egr may be obtained from an equation different from equation (17), or these four parameters are used as arguments. The EGR control valve passage gas flow rate m _egr may be obtained from the map.

Further, according to the present embodiment, since the EGR control valve passage gas flow rate is calculated with high accuracy, the in-cylinder charged fresh air amount calculated based on the EGR control valve passage gas flow rate is also calculated with high accuracy. The fuel injection amount for making the air-fuel ratio the target air-fuel ratio is calculated with high accuracy.
Furthermore, the EGR opening degree may be feedback-controlled using the EGR control valve passage gas flow rate calculated according to the present embodiment so that the EGR control valve passage gas amount becomes the target value.

式（２５）において、Ｍ_c-airは吸気弁が閉弁したときに各気筒内に充填されている新気の量である筒内充填新気量（ｇ）であり、ＤＳＰは内燃機関の排気量（リットル）であり、ＮＣＹＬは気筒数であり、ρ_astdは標準状態（１気圧、２５℃）における空気の密度（約１．２ｇ／リットル）である。 Next, a method for deriving Equation (23) for calculating the exhaust pressure will be described.
The engine load factor (%) is the ratio of the amount of air (g) actually charged into the cylinder to the amount of gas (g) that occupies the maximum volume of the cylinder in the standard state. Obtained from

In Expression (25), _Mc-air is the in _- cylinder charged fresh air amount (g) that is the amount of fresh air that is filled in each cylinder when the intake valve is closed, and DSP is the amount of fresh air in the internal combustion engine. It is the displacement (liter), NCYL is the number of cylinders, and ρ _astd is the density of air (about 1.2 g / liter) in the standard state (1 atm, 25 ° C.).

Referring to Figure 13 showing the relationship between the engine load factor KL and the engine speed NE and the exhaust pressure P _e, the higher the the exhaust pressure P _e engine load factor KL increases, the exhaust pressure engine speed NE increases P _e is made it can be seen that high. Then, the exhaust pressure P _e is primarily because highly dependent on the these engine load factor KL and the engine speed NE, the accuracy of the exhaust pressure obtained from these engine load factor KL and the engine speed NE is relatively high.

Therefore, the exhaust pressure P _e, as in equation (23) can be expressed by a function f ₁ that the engine load factor KL and the engine speed NE as variables (KL, NE). In the first embodiment, this function f ₁ (KL, NE) is stored in advance in the ROM 34 in the form of a map with the engine load factor KL and the engine speed NE as variables, and the engine load factor KL and the engine speed NE are stored. And the exhaust pressure _Pe is calculated from this map.
According to this, since it is not necessary to provide a sensor for detecting the exhaust pressure in the internal combustion engine, it is possible to accurately detect the exhaust pressure while suppressing the cost of the internal combustion engine, and in turn, it passes through the EGR control valve. The gas flow rate m _egr can be calculated with high accuracy.

Also, in an internal combustion engine with an EGR device, when exhaust pressure is detected using a map without using a sensor, usually the engine speed, fuel injection amount, fuel ignition timing, EGR rate (filled in the cylinder) A map having as arguments a number of parameters such as the ratio of the amount of exhaust gas (g) to the amount of gas (g)) had to be used. However, since the engine load factor and the engine speed may be used as parameters by using the engine load factor as a parameter as in this embodiment, the map creation effort can be reduced according to this embodiment. Can be reduced.
Of course, if there is no problem with increasing the cost of the internal combustion engine, the exhaust pressure may be detected using a sensor.

Further, in the case of calculating the exhaust pressure from the function f ₁ described above, the cylinder charge fresh air flow rate _mc-air may be used as a variable instead of the engine speed NE.

Next, a method of deriving the equation for calculating the exhaust temperature T _e (24).
When the amount of heat generated by the exhaust gas newly flowing into the EGR passage 21 is referred to as input heat amount, and the amount of heat released from the EGR passage 21 to the atmosphere is referred to as heat release amount, the EGR control valve passage gas flow rate m _egr increases. When the amount of exhaust gas flowing into the EGR passage 21 increases, the amount of input heat becomes larger than the amount of heat released. Therefore, the exhaust gas temperature increases as the EGR control valve passage gas flow rate m _egr increases.
Further, when the engine speed increases, the temperature of the exhaust gas discharged from the cylinder itself increases.
In other words, the relationship between the EGR control valve passage gas flow rate m _egr and the engine speed NE and the exhaust temperature T _e is as shown in FIG. 14.

Then, the exhaust temperature T _e is mainly because it largely depends on the these EGR control valve passage gas flow rate m _egr and the engine speed NE, these EGR control valve passage gas flow rate m _egr and the engine speed NE The accuracy of the exhaust temperature required is relatively high.
Therefore, the exhaust temperature T _e, as in equation (24), expressed by the EGR control valve passage gas flow rate m _egr and the engine speed function has a NE as a variable f ₂ (m _egr, NE). In the first embodiment, this function f ₂ (m _egr , NE) is stored in advance in the ROM 34 in the form of a map using the EGR control valve passage gas flow rate m _egr and the engine speed NE as variables, and passed through the EGR control valve. The exhaust gas temperature _Te is calculated from the gas flow rate m _egr , the engine speed NE, and this map.

According to this, since it is not necessary to provide a sensor for detecting the exhaust gas temperature in the internal combustion engine, it is possible to detect the exhaust gas temperature accurately while suppressing the cost of the internal combustion engine. The gas flow rate m _egr can be calculated with high accuracy.
Of course, if the increase in the cost of the internal combustion engine is not a problem, the exhaust temperature may be detected using a sensor.

Further, as described above, since the temperature of the exhaust gas discharged from the cylinder changes according to the engine speed NE, the temperature of the exhaust gas can be obtained from a one-dimensional map using the engine speed NE as an argument. .
However, when calculating the EGR control valve passage gas flow rate m _egr at the EGR control valve model M14, the exhaust temperature T _e, rather than using the temperature of the exhaust gas discharged from the cylinders, the EGR control valve 22 upstream of the exhaust gas The EGR control valve passage gas flow rate m _egr can be calculated more accurately by using the temperature.
According to this embodiment, as the exhaust temperature T _e used to calculate the EGR control valve passage gas flow rate m _egr, it means using a temperature of the EGR control valve 22 upstream of the exhaust gas, the EGR control valve passage gas The flow rate m _egr can be calculated with high accuracy.

In addition, when calculating the exhaust gas temperature from the function f ₂ described above, the in-cylinder charged fresh air flow rate _mc-air may be used as a variable instead of the engine speed NE.

  By the way, in the above-described embodiment, it is assumed that it is in steady operation, and therefore the flow rate of exhaust gas flowing into the intake pipe portion (hereinafter referred to as inflowing exhaust gas partial flow rate) is equal to the EGR control valve passage gas flow rate. The flow rate of exhaust gas flowing into the cylinder (hereinafter referred to as in-cylinder inflow exhaust gas flow rate) is also assumed to be equal to the EGR control valve passage gas flow rate.

However, during transient operation, for example, when the EGR opening changes and the EGR control valve passage gas flow rate changes, the inflow exhaust gas partial flow rate and the cylinder inflow exhaust gas flow rate are temporarily set to the EGR control valve passage gas flow rate. Are not equal. That is, these inflow exhaust gas partial flow rate and in-cylinder inflow exhaust gas flow rate follow the change in the EGR control valve passage gas flow rate, but change with a delay.
Therefore, in order to maintain high accuracy of the parameter values calculated in the above-described embodiment during the transient operation, the inflow exhaust gas partial flow rate and the inflow exhaust gas flow rate in the cylinder change in the flow rate of the EGR control valve passage gas. It is necessary to consider that it changes late.

When considering the inflow exhaust gas partial flow rate when the EGR control valve passage gas flow rate changes, as shown in FIG. 15, the inflow exhaust gas partial flow rate m _egr-k is _equal to the EGR control valve passage gas flow rate m _egr. It does not change until a certain time Td elapses after the change, and changes toward the EGR control valve passage gas flow rate m _egr with a primary delay after the certain time Td elapses.

  Here, the fixed time Td is a time required for the exhaust gas that has passed through the EGR control valve 22 to reach the intake pipe portion, and the higher the flow rate of the exhaust gas that has passed through the EGR control valve 22, the more the intake air accordingly. The lower the tube pressure, and hence the shorter the engine speed, the shorter. That is, the fixed time (hereinafter referred to as dead time) Td is a function of the engine speed as shown in FIG.

  On the other hand, the time constant τ of the first-order lag represents the ease of diffusion of the exhaust gas that has passed through the EGR control valve 22, and the higher the flow rate of the exhaust gas that has passed through the EGR control valve 22, the higher the intake pipe. The lower the pressure, the smaller the engine speed. That is, the time constant τ is also a function of the engine speed as shown in FIG.

Therefore, in the present embodiment, the dead time Td1 is stored in advance in the ROM 34 in the form of a map as shown in FIG. 16A, and when the EGR control valve passage gas flow rate changes, it is based on the engine speed NE. The dead time Td1 is calculated, and the inflow exhaust gas partial flow rate is calculated on the assumption that the inflow exhaust gas partial flow rate starts to change toward the EGR control valve passage gas flow rate after the dead time Td1 has elapsed.
In this embodiment, the time constant τ1 is stored in advance in the ROM 34 in the form of a map as shown in FIG. 16B, and when the EGR control valve passage gas flow rate changes, the time constant τ1 is based on the engine speed NE. The time constant τ1 is calculated, and the inflowing exhaust gas partial flow rate is calculated on the assumption that the inflowing exhaust gas partial flow rate changes toward the EGR control valve passage gas flow rate with the time constant τ1 after the dead time Td1 elapses. .

Specifically, represents the calculation period of the EGR control valve passage gas flow rate m _egr at Delta] t (sec), the number of calculations of the EGR control valve passage gas flow rate m _egr and represented by k, the EGR control in the k-th calculation routine After calculating the valve passing gas flow rate m _egr (k), the time constant τ1 is read from the map as shown in FIG. 16B based on the engine speed NE at that time, and EGR is obtained using this time constant τ1. The control valve passage gas flow rate m _egr (k) is subjected to the first order lag processing according to the following equation (26), and the inflow exhaust gas partial flow rate m ′ _egr-k (with the first order lag when the k th calculation routine is performed. k) is calculated.
m ′ _egr-k (k) = m ′ _egr-k (k−1) + Δt / τ1 · (m _egr (k) −m ′ _egr-k (k−1) (26)

Similarly, based on the engine speed NE when the k-th calculation routine is performed, the dead time Td1 is read from the map as shown in FIG. 16A, and the following equation ( 27) The number of wasted routines Idx1 corresponding to the wasted time Td1 is calculated.
Idx1 = Td1 / Δt (27)

Next, by using this waste routine frequency Idx1, the inflow exhaust gas partial flow rate m ' _egr-k (k) is processed in the dead time according to the following equation (28), and the inflow exhaust gas when the kth calculation routine is performed. The gas partial flow rate m _egr-k (k) is calculated.
m _egr-k (k) = m ′ _egr-k (k−Idx1) (28)

The concept of dead time and first-order lag concerning the inflowing exhaust gas partial flow rate is similarly applied to the inflowing exhaust gas flow rate in the cylinder.
That is, the dead time Td2 relating to the in-cylinder inflow exhaust gas flow rate is previously stored in the ROM 34 in the form of a map as shown in FIG. 17A, and when the EGR control valve passage gas flow rate changes, the engine speed NE The in-cylinder inflow exhaust gas flow rate is calculated on the assumption that the in-cylinder inflow exhaust gas flow rate starts to change toward the EGR control valve passage gas flow rate after the dead time Td2 has elapsed. To do.

  Then, the time constant τ2 regarding the in-cylinder inflowing exhaust gas flow rate is stored in advance in the ROM 34 in the form of a map as shown in FIG. 17B, and when the EGR control valve passage gas flow rate changes, the engine speed NE The time constant τ2 is calculated based on the above, and the cylinder inflow exhaust gas flow is assumed to change toward the EGR control valve passage gas flow rate with the time constant τ2 after the dead time Td2 elapses. Calculate the flow rate.

Specifically, after calculating the EGR control valve passage gas flow rate m _egr (k) in the kth calculation routine, the time constant is calculated from the map as shown in FIG. 17B based on the engine speed NE at that time. τ2 is read and the EGR control valve passage gas flow rate m _egr (k) is subjected to the first order lag processing according to the following equation (29) using this time constant τ2 to obtain the first order lag when the kth calculation routine is performed. The in-cylinder inflow exhaust gas flow rate m ′ _egr-egr (k) is calculated.
m ′ _egr-egr (k) = m ′ _egr-egr (k−1) + Δt / τ1 · (m _egr (k) −m ′ _egr-egr (k−1) (29)

Then, based on the engine speed NE when the k-th calculation routine is performed, the dead time Td2 is read from the map as shown in FIG. 17A, and the following equation ( 30) The number of wasted routine times Idx2 corresponding to the wasted time Td2 is calculated.
Idx2 = Td2 / Δt (30)

Next, the cylinder inflow exhaust gas flow rate m ′ _egr-egr (k) is processed in accordance with the following equation (31) using the waste routine frequency Idx, and the cylinder when the kth calculation routine is performed is performed. The inflow exhaust gas flow rate m _egr-egr (k) is calculated.
m _egr-egr (k) = m ′ _egr-egr (k−Idx2) (31)

  Since the distance from the EGR control valve 22 to the cylinder is longer than the distance from the EGR control valve 22 to the intake pipe portion, the dead time Td2 related to the in-cylinder inflow exhaust gas flow rate is greater than the dead time Td1 related to the inflow exhaust gas partial flow rate. The time constant τ2 related to the in-cylinder inflow exhaust gas flow rate tends to be larger than the time constant τ1 related to the inflow exhaust gas partial flow rate.

FIG. 18 shows an example of a routine for calculating the in _- cylinder inflow exhaust gas flow rate m _egr-egr using the first-order lag process and the dead time process as described above.
In the routine shown in FIG. 18, first, in step 10, the EGR control valve passage gas flow rate m _egr (k) is calculated according to the above-described equation (17). In the following step 11, the time constant τ2 is read from the map as shown in FIG. 17B based on the engine speed NE.

Then, in the following step 12, the EGR control valve passage gas flow rate m _egr (k) calculated in step 10 is subjected to a first order lag process, and the cylinder inflow exhaust gas flow rate m ′ _egr-egr (k) with the first order lag is processed. ) Is calculated. That is, here, the above equation (29) is used.
In the following step 13, the dead time Td2 is read from the map as shown in FIG. 17A based on the engine speed NE. Then, in the following step 14, the waste routine count Idx2 is calculated according to the equation (30).

In the following step 15, the cylinder inflow exhaust gas flow rate m ′ _egr-egr (k) calculated in step 12 is subjected to dead time processing according to the equation (31).
In the above-described embodiment, the dead time Td (Td1, Td2) and the time constant τ (τ1, τ2) are calculated using the engine speed NE as a variable. Since the constant τ is substantially proportional to the time required for the crankshaft to rotate 360 °, for example, as shown in FIG. 19, the time T360 ° required for the crankshaft to rotate 360 ° is used as a variable as a dead time. It is preferable to create a map of Td and time constant τ.

Incidentally, the EGR control valve passage gas flow rate m _egr varies depending on various parameters, for example, the number of steps of the EGR control valve 22, that is, the EGR opening degree θ _e , as can be seen from the equation (17).
Therefore, when the EGR opening degree θ _e changes, the EGR opening degree θ _e itself is subjected to dead time processing and delay processing (smoothing process), and the EGR opening degree θ _e after such processing is determined as described above. by inputting the equation (17) that, EGR opening degree θ changes of the EGR control valve passage gas flow rate m _egr with changes in _e to the dead time and tracking lag and the inflow exhaust gas flow rate in the form reflecting the m _egr-k or in-cylinder inflow exhaust gas flow rate m _egr-egr is obtained.

For example, when the EGR opening degree θ _e changes, the dead time Td1 is calculated from the map as shown in FIG. 20A based on the engine speed NE, and after the dead time Td1 has elapsed, the EGR opening is started. degrees start of changing the theta _e, and, also based on the engine speed NE to calculate a better number N1 such from the map such as shown in FIG. 20 (B), EGR opening degree theta _e with this smoothed speed N1 _May be calculated, and the inflowing exhaust gas partial flow rate m _egr-k may be calculated from the equation (17) based on the EGR opening degree θ _e thus _smoothed .

Specifically, when the EGR opening degree θ _e changes from a value θ _e 1 to a value θ _e 2 when the k-th calculation routine is performed, for example, based on the engine speed NE at that time, FIG. The smoothing number N1 is read from the map as shown in (B), the EGR opening θ _e 2 is smoothed according to the following equation (32) using the smoothing number N1, and the k th calculation routine is The EGR opening θ _e ′ (k) with a follow-up delay when it is performed is calculated.
θ _e ′ (k) = {(N1-1) · θ _e (k−1) + θ _e 2} (32)

Similarly, based on the engine speed NE when the k-th calculation routine is performed, a dead time Td1 is read from a map as shown in FIG. 20A, and the following equation ( 33) The number of wasted routine times Idx1 corresponding to the wasted time Td1 is calculated.
Idx1 = Td1 / Δt (33)

Next, the EGR opening θ _e ′ (k) is subjected to dead time processing according to the following equation (34) using the number of waste routines Idx1, and the EGR opening θ _e ( k) is calculated.
θ _e (k) = θ _e '(k−Idx1) (34)
The value calculated by inputting this θ _e (k) into the equation (17) represents the true inflow exhaust gas partial flow rate m _egr-k (k) when the k-th calculation routine is performed. ing.

Of course, the same concept applies to the case where the in-cylinder inflow exhaust gas flow rate is calculated.
That is, when the EGR opening degree θ _e changes, the dead time Td2 is calculated from the map as shown in FIG. 21A based on the engine speed NE, and after the dead time Td2 has elapsed, the EGR opening is started. The degree θ _e is started to be changed, and the smoothing number N2 is calculated from the map as shown in FIG. 21B based on the engine speed NE, and the EGR opening degree θ _e is calculated using the smoothing number N2. _May be calculated, and the in _- cylinder inflow exhaust gas flow rate m _egr-egr may be calculated from the equation (17) based on the EGR opening θ _e thus _smoothed .

Specifically, when the EGR opening degree θ _e changes from, for example, the value θ _e 1 to the value θ _e 2 when the k-th calculation routine is performed, FIG. 21 is based on the engine speed NE at that time. The smoothing number N2 is read from the map as shown in (B), the EGR opening θ _e 2 is smoothed according to the following equation (35) using the smoothing number N2, and the k th calculation routine is The EGR opening degree θ _e ′ (k) with a follow-up delay when it is performed is calculated.
θ _e ′ (k) = {(N2-1) · θ _e (k−1) + θ _e 2} (35)

Similarly, based on the engine speed NE when the k-th calculation routine is performed, the dead time Td2 is read from the map as shown in FIG. 21A, and the following equation ( 36) The number of wasted routines Idx2 corresponding to the wasted time Td2 is calculated according to 36).
Idx2 = Td2 / Δt (36)

Next, the EGR opening θ _e ′ (k) is processed according to the following equation (37) using the waste routine frequency Idx2, and the EGR opening θ _e ( k) is calculated.
θ _e (k) = θ _e '(k-Idx2) (37)
The value calculated by inputting this θ _e (k) into the equation (17) is the true in-cylinder inflow exhaust gas flow rate m _egr-egr (k) when the k-th calculation routine is performed. Represents.
Note that the time constants τ1 and τ2 described above may be used instead of the number of annealing described above.

FIG. 22 shows an example of a routine for calculating the in _- cylinder inflow exhaust gas flow rate m _egr-egr using the _annealing process (following delay process) and the dead time process as described above.
In the routine shown in FIG. 22, first, in step 20, the EGR opening degree θ _e (k) in the k-th calculation routine is read. In the following step 21, the smoothing number N2 is read from the map as shown in FIG. 21B based on the engine speed NE.

In the subsequent step 22, the EGR opening degree θ _e (k) detected in step 20 is subjected to a smoothing process to calculate the EGR opening degree θ ′ _e (k) reflecting the follow-up delay. That is, here, the above equation (35) is used.
In the following step 23, the dead time Td2 is read from the map as shown in FIG. 21A based on the engine speed NE. In the following step 24, the number of waste routines Idx2 is calculated according to the equation (36).
In the following step 25, the EGR opening θ ′ _e (k) reflecting the follow-up delay calculated in step 22 is subjected to dead time processing according to the equation (37). Then, in the following step 26, θ _e (k) calculated in step 25 is input to the above-described equation (17), and the cylinder inflow exhaust gas flow rate m _egr-egr (k) is calculated.

By the way, in the above-described embodiment, the in-cylinder inflow exhaust gas flow rate m _egr-egr is used to calculate the in _- cylinder inflow fresh air amount _{Mc-air. The} fuel injection amount is determined so that the air-fuel ratio becomes the target air-fuel ratio based on _c-air . In other words, in the above-described embodiment, the fuel injection amount is determined on the assumption that the in _- cylinder inflow exhaust gas flow rate m _egr-egr is the same value for each cylinder.
However, the actual in-cylinder inflow exhaust gas flow rate m _egr-egr may vary from cylinder to cylinder depending on the shape of the flow path from the EGR control valve 22 to each cylinder and various other factors. Therefore, in order to determine the fuel injection amount so that the air-fuel ratio becomes the target air-fuel ratio more accurately, it should be considered that the in-cylinder inflow exhaust gas flow rate m _egr-egr differs for each cylinder.

Therefore, the ratio of the actual in-cylinder inflow exhaust gas flow rate in each cylinder to the in _- cylinder inflow exhaust gas flow rate m _egr-egr calculated in the above-described embodiment is obtained in advance by experiments or the like as a distribution coefficient, or during engine operation The in-cylinder inflow exhaust gas after the distribution coefficient is reflected by multiplying the in _- cylinder inflow exhaust gas flow rate m _egr-egr calculated in the above-described embodiment by the distribution coefficient every time a specific condition is satisfied. The fuel injection amount may be determined for each cylinder using the flow rate m _egr-egr .
If this method is adopted, the air-fuel ratio becomes the target air-fuel ratio more accurately.

In addition, the distribution coefficient regarding each cylinder is calculated | required beforehand by experiment etc. as follows, or is calculated | required during engine operation, for example.
That is, in the operating state where the EGR rate (cylinder charged exhaust gas amount / cylinder charged gas amount) is the largest, the same amount of fuel is supplied to each cylinder while the EGR opening is kept constant and the throttle opening is kept constant. At this time, the air-fuel ratio of the exhaust gas discharged from each cylinder is detected, and the air-fuel ratio in each cylinder is estimated based on the detected value of the air-fuel ratio of this exhaust gas.

Here, regarding a cylinder whose air-fuel ratio is estimated to be lean, the cylinder is filled with air more than expected, and therefore the cylinder is not filled with exhaust gas as expected. . In this case, the distribution coefficient for this cylinder is less than 1.0. Specifically, the in-cylinder inflow exhaust gas flow rate related to this cylinder is back-calculated based on the estimated air-fuel ratio, and the in-cylinder in-cylinder calculated by this back-calculation with the in-cylinder inflow exhaust gas flow rate calculated in the above-described embodiment. By dividing the inflow exhaust gas flow rate, the distribution coefficient for this cylinder is calculated.
On the other hand, for the cylinders estimated to be rich in the air-fuel ratio, the distribution coefficient can be calculated in the same way, and in this case, the distribution coefficient is larger than 1.0.

FIG. 23 shows an example of a routine for calculating the cylinder inflow exhaust gas flow rate m _egr-egr in consideration of the distribution coefficient for each cylinder as described above. The following routine is applied to an internal combustion engine having four cylinders.
In the routine shown in FIG. 23, first, in step 30, the EGR control valve passage gas flow rate m _egr (k) is calculated according to the above-described equation (17). In the following step 31, the time constant τ2 is read from the map as shown in FIG. 17B based on the engine speed NE.

In the subsequent step 32, the EGR control valve passage gas flow rate m _egr (k) calculated in step 30 is subjected to a first-order lag process, and the cylinder inflow exhaust gas flow rate m ′ _egr-egr (k ) Is calculated. That is, here, the above equation (29) is used.
In the following step 33, the dead time Td2 is read from the map as shown in FIG. 17A based on the engine speed NE. In subsequent step 34, the number of waste routines Idx2 is calculated according to the equation (30).

In the following step 35, the in-cylinder inflow exhaust gas flow rate m ′ _egr-egr (k) calculated in step 32 is subjected to dead time processing according to the equation (31).
Finally, in step 36, the in-cylinder inflow exhaust gas flow rate m _egr-egr (k) calculated in step 35 is multiplied by distribution coefficients K1 to K4 for each cylinder according to the following equation (38). In-cylinder inflow exhaust gas flow rate m _egr-egr (k) (1) to (4) for each cylinder is calculated.
m _egr-egr (k) (1) = m _egr-egr (k) ・ K1
m _egr-egr (k) (2) = m _egr-egr (k) ・ K2
m _egr-egr (k) (3) = m _egr-egr (k) ・ K3
m _egr-egr (k) (4) = m _egr-egr (k) ・ K4 (38)

In equation (38), m _egr-egr (k) (1) is the in-cylinder inflow exhaust gas flow rate for the first cylinder, K1 is the distribution coefficient for the first cylinder, and m _egr-egr (k) (2) is the cylinder inflow exhaust gas flow rate for the second cylinder, K2 is the distribution coefficient for the second cylinder, and m _egr-egr (k) (3) is the cylinder inflow exhaust gas flow rate for the third cylinder. Yes, K3 is the distribution coefficient for the third cylinder, m _egr-egr (k) (4) is the in-cylinder inflow exhaust gas flow rate for the fourth cylinder, and K4 is the distribution coefficient for the fourth cylinder.

By the way, in the above-described embodiment, considering that the actual in-cylinder inflow exhaust gas flow rate m _egr-egr varies from cylinder to cylinder due to various factors such as the shape of the flow path from the EGR control valve 22 to each cylinder. The distribution coefficient for each cylinder is obtained, and the in-cylinder inflow exhaust gas flow rate m _egr-egr is obtained for each cylinder. For example, the flow path length from the EGR control valve 22 to each cylinder is very long. When the flow path resistance is very large, the dead time, the time constant, and the number of annealing are different.

Therefore, as in the above-described embodiment, not only the distribution coefficient is reflected in the obtained in-cylinder inflow exhaust gas flow rate m _egr-egr but also the distribution coefficient is reflected in the dead time, the time constant, or the number of annealing. You may let them. That is, more specifically, in the above-described embodiment, only the obtained in-cylinder inflow exhaust gas flow rate m _egr-egr is multiplied by the distribution coefficient, but the _dead time and time constant, or the annealing The number may be multiplied by the same distribution factor for each cylinder.
According to this, the cylinder inflow exhaust gas flow rate m _egr-egr is calculated more accurately for each cylinder.

FIG. 24 shows an example of a routine for calculating the cylinder inflow exhaust gas flow rate m _egr-egr in each cylinder by reflecting the distribution coefficient for each cylinder in the dead time and the time constant.
In the routine shown in FIG. 24, first, in step 40, the EGR control valve passage gas flow rate m _egr is calculated from the equation (17). In the following step 41, the time constant τ2 is read from the map shown in FIG. 17B based on the engine speed NE. In the subsequent step 42, the dead time Td2 is read from the map shown in FIG. 17A based on the engine speed NE. In the following step 43, 1 is input to the parameter cyl indicating the cylinder number. In this case, cyl = 1, and the parameter cyl indicates the first cylinder.

In the following step 44, it is determined whether or not the parameter cyl is less than or equal to the total number of cylinders Ncyl (cyl ≦ Ncyl). When it is determined that cyl ≦ Ncyl, the routine proceeds to step 45, where the EGR control valve passage gas flow rate m _egr (k) in the kth calculation routine from the above equation (17) is obtained. Calculated.

In the following step 46, the time constant τ2 (cyl) related to the corresponding cylinder from the following equation (39), that is, when the parameter cyl is 1, the time constant τ2 (1) regarding the first cylinder and when the parameter cyl is 2. A time constant τ2 (2) for the second cylinder is calculated.
τ2 (cyl) = K (cyl) · τ2 (39)
Here, K (cyl) is a distribution coefficient for each cylinder. For example, when the parameter cyl is 1, the distribution coefficient is K (1) for the first cylinder, and when the parameter cyl is 2, the distribution is for the second cylinder. The coefficient K (2).

In subsequent step 47, the EGR control valve passage gas flow rate m _egr (k) is subjected to primary processing according to the following equation (40), and the in-cylinder inflow exhaust gas flow rate m ′ _egr-egr ( cyl) (k) is calculated.
m ′ _egr-egr (cyl) (k) = m ′ _egr-egr (cyl) (k−1) + Δt / τ2 (cyl) · (m _egr (cyl) (k) −m ′ _egr-egr (cyl) (K-1)) (40)
Here, Δt is a calculation cycle of the EGR control valve passage gas flow rate.

In the following step 48, the number of wasted routines Idx2 corresponding to the wasted time for the corresponding cylinder is calculated according to the following equation (41).
Idx2 = K (cyl) · Td2 / Δt (41)
In the following step 49, the in-cylinder inflow exhaust gas flow rate m _egr-egr (cyl) (k) related to the corresponding cylinder in the k-th calculation routine is calculated from the following equation (42) in the form of dead time processing.
m _egr-egr (cyl) (k) = m ′ _egr-egr (cyl) (k−Idx2) (42)

  In the following step 50, the parameter cyl is increased by 1, and the routine returns to step 44. Therefore, if the parameter cyl is 1 when the routine reaches step 50, the parameter cyl is set to 2 in step 50, and steps 44 to 49 are repeated.

If the parameter cyl is 4 when the routine arrives at step 50, the parameter cyl is set to 5 at step 50, and when the routine returns to step 44, at step 44, cyl> The routine proceeds to step 51 after being determined as Ncyl. That is, when it is determined in step 44 that cyl> Ncyl, the cylinder inflow exhaust gas flow rate m _egr-egr (cyl) (k) is calculated for all the cylinders.
In step 51, zero is input to the parameter cyl, that is, the parameter cyl is cleared.
This routine can also be used as a routine for calculating the inflow exhaust gas partial flow rate m _egr-k flowing into each intake branch pipe 13 by changing Td2 to Td1 and τ2 to τ1.

By the way, an internal combustion engine provided with a valve timing changing mechanism for changing the closing timing of the intake valve is known. For example, in one valve timing changing mechanism, the closing timing of the intake valve can be changed after the intake bottom dead center.
When the intake valve closes after intake bottom dead center, part of the gas that has once flowed into the cylinder is discharged from the cylinder into the intake passage by the piston in the cylinder. The amount of gas discharged from the cylinder into the intake passage (hereinafter also referred to as blow-back gas) varies depending on the timing at which the intake valve closes after intake bottom dead center. Increases as the valve closes at a later timing after the intake bottom dead center.

That is, in an internal combustion engine having a valve timing mechanism that changes the closing timing of the intake valve after intake bottom dead center, when the closing timing of the intake valve is changed by this valve timing mechanism, the amount of blowback gas changes. To do.
Here, considering the diffusion of exhaust gas in the intake passage when gas is exhausted from the cylinder to the intake passage in this way, the greater the amount of blowback gas, the more the exhaust gas diffuses in the intake passage. Tend to be suppressed. That is, as the amount of blowback gas increases, the diffusion of the exhaust gas that has passed through the EGR control valve 22 and entered the intake passage tends to be suppressed.
That is, this means that the time constant τ2 increases as the amount of blowback gas increases, that is, as the closing timing of the intake valve is delayed.

Further, when considering the EGR passage 21, the pressure in the EGR passage 21 is significantly higher than the pressure in the intake passage. Therefore, even if the gas is discharged from the cylinder to the intake passage, the gas enters the EGR passage 21. There is no inflow. Therefore, even if gas is discharged from the cylinder to the intake passage, the time taken for the exhaust gas that has passed through the EGR control valve 22 to reach the cylinder does not change much.
That is, this means that the above-described dead time Td2 does not change so much regardless of the amount of blowback gas, that is, regardless of the closing timing of the intake valve.

  In view of this, the internal combustion engine in which the valve closing timing of the intake valve can be changed after the intake bottom dead center in consideration of the change in the time constant τ2 described above in accordance with the valve closing timing of the intake valve. In the above-described embodiment, the dead time Td2 and the time constant τ2 may be determined according to the intake valve closing timing with the relationship shown in FIGS. 25 (A) and (B). .

As shown in FIG. 25A, the dead time Td2 is used here regardless of the intake valve closing crank angle CA corresponding to the closing timing of the intake valve at any angle after the intake bottom dead center BDC. Is constant.
On the other hand, as shown in FIG. 25B, here, the time constant τ2 increases as the intake valve closing crank angle CA becomes a slower angle after the intake bottom dead center BDC.

As described above, since the dead time Td2 and the time constant τ2 also depend on the engine speed NE, the dead time Td2 and the time constant τ2 are determined as a function of the engine speed NE and the intake valve closing timing. You may make it do.
For example, in this case, FIGS. 17A and 17B show the dead time Td2 and the time constant τ2 when the closing timing of the intake valve is the intake bottom dead center, with the engine speed NE as a variable. Such a map is obtained in advance and stored in the ROM 34.

  The dead time is the dead time Td2 obtained from the map shown in FIG. 17A based on the engine speed NE, and the dead time Td2 shown in FIG. 26A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map shown is used. In this case, as can be seen from FIG. 26 (A), the correction coefficient Ktd is constant at 1.0 regardless of the intake valve closing crank angle CA. The dead time Td2 obtained from the map shown in FIG.

  On the other hand, the time constant is a time constant τ2 obtained from the map shown in FIG. 17B based on the engine speed NE, and the time constant shown in FIG. 26B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used. In this case, as can be seen from FIG. 26B, the correction coefficient Kτ is 1.0 when the intake valve closing crank angle CA is at the intake bottom dead center BDC, and the intake valve closing crank angle CA is the intake air. The slower the lower dead center BDC, the larger the value.

  Note that in the internal combustion engine in which the intake valve closing timing can be changed after the intake bottom dead center, the above-described dead time has the same relationship as that shown in FIGS. 25 (A) and (B). Td1 and time constant τ1 may be determined. In this case, the dead time Td1 and the time constant τ1 are smaller than the dead time Td2 and the time constant τ2.

Further, the dead time Td1 and the time constant τ1 may be determined as a function of the engine speed NE and the intake valve closing timing.
For example, in this case, as shown in FIGS. 16A and 16B, the dead time Td1 and the time constant τ1 when the intake valve closing timing is the intake bottom dead center are used with the engine speed NE as a variable. It is obtained in advance in the form of a simple map and stored in the ROM 34.

  The dead time is the dead time Td1 obtained from the map shown in FIG. 16A based on the engine speed NE, and the dead time Td1 shown in FIG. 26A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map shown is used. In this case, as can be seen from FIG. 26 (A), the correction coefficient Ktd is constant at 1.0 regardless of the intake valve closing crank angle CA. The dead time Td1 obtained from the map shown in FIG.

  On the other hand, the time constant is a time constant τ1 obtained from the map shown in FIG. 16B based on the engine speed NE, and the time constant shown in FIG. 26B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used. In this case, as can be seen from FIG. 26B, the correction coefficient Kτ is 1.0 when the intake valve closing crank angle CA is at the intake bottom dead center BDC, and the intake valve closing crank angle CA is the intake air. The slower the lower dead center BDC, the larger the value.

By the way, in an internal combustion engine equipped with a valve timing mechanism that changes the closing timing of the intake valve before the intake bottom dead center, when the closing timing of the intake valve is changed by this valve timing mechanism, the amount of blowback gas is almost the same. Although it is zero, the total amount of gas charged in the cylinder varies.
Here, when the amount of blowback gas is almost zero in this way, even if the closing timing of the intake valve changes before the intake bottom dead center, the change in the closing timing of the intake valve does not occur in the intake passage. Does not affect the diffusion of exhaust gas.
That is, this means that the above-mentioned time constant τ2 is constant regardless of the closing timing of the intake valve.

  On the other hand, when considering the time taken for the exhaust gas to reach the cylinder from the EGR control valve 22 when the total gas amount filled in the cylinder changes, the smaller the total gas amount filled in the cylinder, the smaller the total gas amount filled in the cylinder. This time tends to be longer.

  That is, this means that the above-mentioned dead time Td2 becomes longer as the total amount of gas filled in the cylinder is smaller, that is, as the intake valve closing timing is earlier in the range before the intake bottom dead center. Means.

In the internal combustion engine in which the closing timing of the intake valve can be changed before the intake bottom dead center in consideration of the change in the dead time Td2 according to the closing timing of the intake valve. In the above-described embodiment, the dead time Td2 and the time constant τ2 may be determined according to the closing timing of the intake valve with the relationship as shown in FIGS. 27 (A) and (B).
As shown in FIG. 27A, the dead time Td2 increases as the intake valve closing crank angle CA becomes an earlier angle before the intake bottom dead center BDC.

On the other hand, as shown in FIG. 27B, the time constant τ2 is constant no matter what the intake valve closing crank angle CA is before the intake bottom dead center BDC.
As described above, since the dead time Td2 and the time constant τ2 also depend on the engine speed NE, the dead time Td2 and the time constant τ2 are determined as a function of the engine speed NE and the intake valve closing timing. You may make it do.
For example, in this case, FIGS. 17A and 17B show the dead time Td2 and the time constant τ2 when the closing timing of the intake valve is the intake bottom dead center, with the engine speed NE as a variable. Such a map is obtained in advance and stored in the ROM 34.

  The dead time is shown as dead time Td2 obtained from the map shown in FIG. 17A based on the engine speed NE, and as shown in FIG. 28A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map is used. In this case, as can be seen from FIG. 28A, the correction coefficient Ktd is 1.0 when the intake valve closing crank angle CA is at the intake bottom dead center BDC, and the intake valve closing crank angle CA is the intake air. The value becomes larger as it becomes earlier than the bottom dead center BDC.

  On the other hand, the time constant is a time constant τ2 obtained from the map shown in FIG. 17B based on the engine speed NE, and the time constant shown in FIG. 28B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used. In this case, as can be seen from FIG. 28 (B), the correction coefficient Kτ is constant at 1.0 irrespective of the intake valve closing crank angle CA. The time constant τ2 obtained from the map shown in FIG.

  In the internal combustion engine in which the intake valve closing timing can be changed before the intake bottom dead center, the above-described dead time has the same relationship as that shown in FIGS. 27 (A) and 27 (B). Td1 and time constant τ1 may be determined. In this case, the dead time Td1 and the time constant τ1 are smaller than the dead time Td2 and the time constant τ2.

Further, the dead time Td1 and the time constant τ1 may be determined as a function of the engine speed NE and the intake valve closing timing.
For example, in this case, as shown in FIGS. 16A and 16B, the dead time Td1 and the time constant τ1 when the intake valve closing timing is the intake bottom dead center are used with the engine speed NE as a variable. It is obtained in advance in the form of a simple map and stored in the ROM 34.

  The dead time is the dead time Td1 obtained from the map shown in FIG. 16A based on the engine speed NE, and the dead time Td1 shown in FIG. 28A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map shown is used. In this case, as can be seen from FIG. 28A, the correction coefficient Ktd is 1.0 when the intake valve closing crank angle CA is at the intake bottom dead center BDC, and the intake valve closing crank angle CA is the intake air. The faster the bottom dead center, the larger the value.

  On the other hand, the time constant is a time constant τ1 obtained from the map shown in FIG. 16B based on the engine speed NE, and the time constant shown in FIG. 28B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used. In this case, as can be seen from FIG. 28 (B), the correction coefficient Kτ is constant at 1.0 irrespective of the intake valve closing crank angle CA. The time constant τ1 obtained from the map shown in FIG.

Incidentally, an internal combustion engine having a valve timing mechanism that can change the intake valve closing timing across the intake bottom dead center is known. Such a valve timing mechanism employs, for example, a mechanism that electromagnetically opens or closes the intake valve.
If the intake valve closing timing can be changed across the intake bottom dead center in this way, is the intake valve closing timing changed before the intake bottom dead center or after the intake bottom dead center? The above-described dead time Td2 and time constant τ2 differ.

  Here, when the intake valve closing timing is changed before the intake bottom dead center, the dead time Td2 and the time constant τ2 show the tendency as described in relation to FIG. On the other hand, when the intake valve closing timing is changed after the intake bottom dead center, the dead time Td2 and the time constant τ2 show the tendency as described in relation to FIG.

  Therefore, in the internal combustion engine in which the intake valve closing timing can be changed across the intake bottom dead center, the relationship shown in FIGS. 29A and 29B in the above-described embodiment. Therefore, the dead time Td2 and the time constant τ2 may be determined according to the intake valve closing timing.

As shown in FIG. 29A, here, in the region before the intake bottom dead center BDC, the earlier the intake valve closing crank angle CA is, the larger the dead time Td2 is. In the area after the dead center BDC, the dead time Td2 is constant regardless of the intake valve closing crank angle CA.
On the other hand, as shown in FIG. 29B, here, in the region before the intake bottom dead center BDC, the time constant τ2 is constant regardless of the intake valve closing crank angle CA, and the intake air In the region after the bottom dead center BDC, the time constant τ2 increases as the intake valve closing crank angle CA becomes slower.
As described above, since the dead time Td2 and the time constant τ2 also depend on the engine speed NE, the dead time Td2 and the time constant τ2 are determined as a function of the engine speed NE and the intake valve closing timing. You may make it do.

  For example, in this case, as shown in FIGS. 17A and 17B, the dead time Td2 and the time constant τ2 when the intake valve closing timing is the intake bottom dead center are used with the engine speed NE as a variable. It is obtained in advance in the form of a simple map and stored in the ROM 34.

  The dead time is the dead time Td2 obtained from the map shown in FIG. 17A based on the engine speed NE, and the dead time Td2 shown in FIG. 30A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map shown is used. Here, in the map of FIG. 30A, the correction coefficient Ktd increases as the intake valve closing crank angle CA increases in the region before the intake bottom dead center BDC, and in the region after the intake bottom dead center BDC, Regardless of the intake valve closing crank angle CA, the correction coefficient Ktd is constant at 1.0.

On the other hand, the time constant is a time constant τ2 obtained from the map shown in FIG. 17B based on the engine speed NE, and the time constant shown in FIG. 30B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used. Here, in the map of FIG. 30B, in the region before the intake bottom dead center BDC, the correction coefficient Kτ is constant at 1.0 regardless of the intake valve closing crank angle CA, and the intake In the area after the dead center BDC, the correction coefficient Kτ increases as the intake valve closing crank angle CA becomes slower.
Note that in the internal combustion engine in which the intake valve closing timing can be changed across the intake bottom dead center, the above-described waste is caused by the same relationship as that shown in FIGS. 29 (A) and 29 (B). The time Td1 and the time constant τ1 may be determined. In this case, the dead time Td1 and the time constant τ1 are smaller than the dead time Td2 and the time constant τ2.

Further, the dead time Td1 and the time constant τ1 may be determined as a function of the engine speed NE and the intake valve closing timing.
For example, in this case, as shown in FIGS. 16A and 16B, the dead time Td1 and the time constant τ1 when the intake valve closing timing is the intake bottom dead center are used with the engine speed NE as a variable. It is obtained in advance in the form of a simple map and stored in the ROM 34.

  The dead time is the dead time Td1 obtained from the map shown in FIG. 16A based on the engine speed NE, and the dead time Td1 shown in FIG. 30A based on the intake valve closing crank angle CA. A value obtained by multiplying the correction coefficient Ktd obtained from the map shown is used.

  On the other hand, the time constant is a time constant τ1 obtained from the map shown in FIG. 16B based on the engine speed NE, and the time constant shown in FIG. 30B based on the intake valve closing crank angle CA. A value multiplied by a correction coefficient Kτ obtained from the map shown is used.

Incidentally, in the above-described embodiment, the EGR control valve passage gas flow rate m _egr is calculated using a calculation formula, that is, the formulas (17) and (18). The inventor of the present application uses such a calculation formula. Even if not, a method of calculating the EGR control valve passage gas flow rate m _egr relatively easily was devised. Therefore, in the above-described embodiment, instead of using the equations (17) and (18), the EGR control valve passage gas flow rate may be calculated using the method described below.

Cylinder charged gas amount M _c is known to be expressed by a linear function of the intake pipe pressure P _m of when the intake valve 6 is closed. That is, according to theory and empirical rule, the in-cylinder charged gas amount _Mc is proportional to the in-cylinder pressure when the intake valve 6 is closed, and this in-cylinder pressure is the mixed gas pressure upstream of the intake valve 6, that is, It almost coincides with the intake pipe pressure P _m .
When EGR gas is not supplied, only fresh air is filled into the cylinder, and therefore, the cylinder fresh charge amount M _{c-air at} this time, and therefore the engine load factor KL, is expressed as a linear function equation of the intake pipe pressure P _m. It can be expressed as That is, the engine load factor KL can be obtained easily and accurately.

However, the situation is completely different when EGR gas is supplied, and not only fresh air but also EGR gas is filled in the cylinder. Therefore, in the conventional in-cylinder charging fresh air amount M _c-air or, is the fact was thought absolutely impossible to represent the engine load factor KL by a linear function expression of the intake pipe pressure P _m.
If it is possible to represent the in-cylinder charged EGR gas amount M _c-egr by a linear function of the intake pipe pressure P _m, it can represent a cylinder charged gas amount M _c by a linear function of the intake pipe pressure P _m , considering that cylinder charged gas amount M _c is the sum of the cylinder charging fresh air amount M _c-air and cylinder charged EGR gas amount M _c-egr, cylinder when the EGR gas is being supplied The internal charge fresh air amount M _c-air or the engine load factor KL can be expressed by a linear function expression of the intake pipe pressure P _m .

However, in the conventional, cylinder charged EGR gas amount M _c-egr is also had been considered can not be represented by a linear function expression of the intake pipe pressure P _m. This will be described with reference to FIGS.
First, as described above, the EGR control valve passage gas flow rate m _egr (g / sec) is expressed by Expression (17), and the function Φ (P _m / P _e ) is expressed by Expression (18).

Here, the exhaust pressure P _e in order to simplify the calculation when the atmospheric pressure P _a, EGR control valve passage gas flow rate m _egr represented by the formula (17) is as shown in FIG 31. That is, the EGR control valve passage gas flow rate m _egr is maintained substantially constant when the intake pipe pressure P _m is small, and is nonlinear with respect to the intake pipe pressure P _m as indicated by NR in FIG. 31 when the intake pipe pressure P _m increases. It decreases toward the atmospheric pressure P _a while showing sex. Note that the nonlinear portion NR is due to P _e / √T _e parts and functions Φ (P _m / P _e) of the formula (17).

Therefore, it has been considered that the EGR control valve passage gas flow rate m _egr , especially the nonlinear portion NR cannot be expressed by a linear function expression of the intake pipe pressure P _m . However, if a considerably large number of linear function equations are used, it is considered that the EGR control valve passage gas flow rate m _egr can be expressed by a linear function equation of the intake pipe pressure P _m . However, in this case, the engine load factor KL can no longer be simply obtained.

However, according to the inventors of the present application, the EGR control valve passage gas flow rate m _egr can be expressed by two linear function expressions of the intake pipe pressure P _m , and therefore, the in-cylinder charged fresh air amount M _c-air , or is the it has been found that it is possible to represent the engine load factor KL by two first-order function equation of the intake pipe pressure P _m.
That is, first, as shown in FIG. 32, the exhaust temperature T _e is to increase the intake pipe pressure P _m, the exhaust pressure P _e increases significantly than increases, as a result, P _e / √T _e the it can be expressed by a linear function expression of the intake pipe pressure P _m.

The function Φ (P _m / P _e ) can also be expressed by a linear function expression of the intake pipe pressure P _m . This will be described with reference to FIG. Considering that the exhaust pressure Pe is not maintained at _a constant atmospheric pressure _Pa but varies according to the intake pipe pressure P _m , as shown in FIG. 33 (A), the intake pipe pressure P _m The function Φ (P _m / P _e ) when is P _m 1 is not on the curve C _a that converges to the atmospheric pressure P _a , but is on the curve C 1 that converges to the exhaust pressure P _e 1, which is plotted (○) Similarly, Φ (P _m / P _e ) when P _m = P _m 2 (> P _m 1) is on the curve C2 that converges to the exhaust pressure P _e 2 (> P _e 1), and P _m = Φ (P _m / P _e ) when P _m 3 (> P _m 2) is on a curve C3 that converges to the exhaust pressure P _e 3 (> P _e 2).

The plot obtained in this way can be connected by a straight line L2, as shown in FIG. Therefore, the function Φ (P _m / P _e) is a linear function of the intake pipe pressure P _m corresponding to the straight line L1 when the intake pipe pressure P _m is small, corresponds to the straight line L2 when the intake pipe pressure P _m is greater intake can be represented by a linear function of the pipe pressure P _m, it will be able to thus to represent in two linear functions of the intake pipe pressure P _m. That is, the EGR control valve passage gas flow rate m _egr can be expressed by two linear function expressions of the intake pipe pressure P _m .

Here, during steady operation, the EGR control valve passage gas flow rate m _egr (g / sec) and the in-cylinder inflow EGR gas flow rate m _c-egr (g / sec) are equal to each other. The cylinder filling EGR gas amount M _c-egr is _{calculated by adding the} time required for one intake stroke of each cylinder to the cylinder inflow EGR gas flow rate m _c-egr , that is, the time required for one cycle of the internal combustion engine. It is obtained by multiplying the time ΔT (sec) divided by (M _c-egr = m _c-egr · ΔT).
Then, the cylinder filling EGR gas amount M _c-egr during steady operation can be expressed by a linear function expression of the intake pipe pressure P _m .

Accordingly, the in-cylinder charged fresh air amount M _c-air or the engine load factor KL during steady operation and when EGR gas is supplied can be expressed by two linear function expressions of the intake pipe pressure P _m. It turns out that.

Assuming that the engine load factor KL when the EGR gas is supplied is KLon, FIG. 34 shows a state during steady operation (for example, when the engine speed NE and the EGR opening degree θ _e are held constant). An example of two linear function expressions of the intake pipe pressure P _m representing the engine load factor KLon is shown. As shown in FIG. 34, the engine load factor KLon, the slope is continuous in different and the connection point CP with each other is represented by two first-order function equation of the intake pipe pressure P _m. That is, when the intake pipe pressure P _m is small, the engine load factor KLon is expressed by a linear function equation of the gradient e1, and when the intake pipe pressure P _m is high, the engine load factor KLon is expressed by a linear function equation of the gradient e2. .

Here, assuming that the gradients of the two linear function equations are e1 and e2, respectively, and the intake pipe pressure and the engine load factor at the connection point CP are d and r, respectively, these two linear function equations can be expressed by the following equations. .
KLon = e1 · (P _m −d) + r... P _m ≦ d
KLon = e 2 · (P _m −d) + r... P _m > d
These are collectively expressed as the following equation (43).
KLon = e · (P _m −d) + r (43)
e = e1 ... P _m ≦ d
e = e2 ... P _m> d

In this embodiment, two linear functions of the intake pipe pressure P _m representing the engine load factor KLon at the time of steady state operation is stored in advance in the ROM34 in the form shown in equation (43). In this way, two linear function expressions can be represented by three parameters e, d, and r. That is, it is possible to reduce the number of parameters necessary to express two linear function expressions.
Each parameter e, d, r of this formula (43) is calculated based on the following formula.
e1 = e1 ^* · ktha
e2 = e2 ^* · ktha
d = d ^* · ktha · kpa
r = r ^* · ktha · kpa
Here, e1 ^* , e2 ^* , d ^* , and r ^* are the gradient, the intake pipe pressure and the engine load factor at the connection point, respectively, when the engine ambient environment is a predetermined reference environment. Any state may be used as the reference environment state, but in this embodiment, the standard state (1 atm, 25 ° C.) is used as the reference environment state.

On the other hand, ktha represents an atmospheric temperature correction coefficient, and kpa represents an atmospheric pressure correction coefficient. The atmospheric temperature correction coefficient ktha is for correcting each parameter e1 ^* , e2 ^* , d ^* , r ^* in the reference environment state according to the actual atmospheric temperature detected by the atmospheric temperature sensor 44, and is corrected. When there is no need, 1.0 is set. The atmospheric pressure correction coefficient kpa is used to correct the parameters d ^* and r ^* in the reference environment state according to the actual atmospheric pressure detected by the atmospheric pressure sensor 45, and when correction is not necessary. 1.0.

Therefore, considering that the atmospheric temperature correction coefficient ktha or the atmospheric pressure correction coefficient kpa is a representative value representative of the actual engine ambient environment condition, the reference environment condition is based on the representative value representative of the actual engine environment environment condition. This means that the parameters e1 ^* , e2 ^* , d ^* , and r ^* are corrected. Alternatively, it can be considered that the engine load factor KLon in the reference environmental state is corrected based on the representative value representative of the actual environmental environment surrounding the engine.

On the other hand, in consideration of the fact that the opening cross-sectional area A _e of the EGR control valve 22 in equation (17) depends on the EGR opening θ _e and the engine charging efficiency depends on the engine speed NE, ^{e * (e1 *, e2 *} ), d *, and the r ^* is set according to the EGR opening theta _e or the engine speed NE.

Specifically, as shown in FIG. 35A, the gradient e1 ^* increases as the engine speed NE increases when the engine speed NE is low, and the engine speed NE when the engine speed NE is high. It becomes smaller as NE becomes higher, and further becomes larger as EGR opening degree θ _e becomes larger. Further, as shown in FIG. 35B, the gradient e2 ^* increases as the engine speed NE increases when the engine speed NE is low, and the engine speed NE increases when the engine speed NE is high. It becomes smaller as the EGR opening degree θ _e becomes larger. These gradients e1 ^*, e2 ^* are obtained by experiments in advance, advance in the ROM34 in the form of a map shown in FIG. 35 (C) and (D) as a function of the respective engine speed NE and EGR opening theta _e It is remembered.

On the other hand, the intake pipe pressure d ^{* at} the connection point CP decreases as the engine speed NE increases, as shown in FIG. The intake pipe pressure d ^{* at the} connection point CP is also obtained in advance by experiments and is stored in advance in the ROM 34 in the form of a map shown in FIG. 36 as a function of the engine speed NE.

Further, as shown in FIG. 37A, the engine load factor r ^{* at} the connection point CP increases as the engine speed NE increases when the engine speed NE is low, and the engine load factor r ^* when the engine speed NE is high. It decreases as the rotational speed NE increases, and further decreases as the EGR opening degree θ _e increases. Has been determined by the engine load factor r ^* also advance by experiments at the connection point CP, is stored in advance in the ROM34 in the form of a map shown in FIG. 37 (B) as a function of the engine speed NE and the EGR opening theta _e Yes.

Therefore, generally speaking, with respect to a plurality of different EGR openings θ _e , the in-cylinder charged fresh air amount M _c-air during steady operation or the intake pipe pressure P _m representing the engine load factor KLon is 2 One linear function expression is obtained and stored in advance. Further, two linear function expressions of the in-cylinder charged fresh air amount M _c-air during steady operation or the intake pipe pressure P _m representing the engine load factor KLon are obtained in advance for a plurality of different engine speeds NE. It also means that it is remembered.

Figure 38 is in and various EGR opening theta _e at a constant engine speed NE, it shows an example of two linear functions of the intake pipe pressure P _m representing the engine load factor KLon at the time of steady operation. The broken line in FIG. 38 indicates the engine load factor KLoff when EGR gas is not supplied, that is, when the EGR opening degree θ _e is zero.

On the other hand, as described above, can represent the engine load factor KLoff when the EGR gas is not being supplied by a linear function expression of the intake pipe pressure P _m. Figure 39 is a steady operation timing (e.g., function speed NE is when it is held constant) are shown an example of two linear functions of the intake pipe pressure P _m representing the engine load factor KLoff of. In the present embodiment, as shown in FIG. 39, the engine load factor KLoff is the slope is continuous in different and the connection point CP with each other is represented by two first-order function equation of the intake pipe pressure P _m. That is, when the intake pipe pressure P _m is small, the engine load factor KLoff is expressed by a linear function equation of the gradient a1, and when the intake pipe pressure P _m is high, the engine load factor KLoff is expressed by a linear function equation of the gradient a2.

Here, assuming that the slopes of the two linear function equations are a1 and a2, respectively, and the intake pipe pressure and the engine load factor at the connection point CP are d and c, respectively, these two linear function equations can be expressed by the following equations. .
_{KLoff = a1 · (P m -d} ) + c ... P m ≦ d
KLoff = a 2 · (P _m −d) + c... P _m > d
These are collectively expressed as the following formula (44).
KLoff = a · (P _m −d) + c (44)
a = a1 ... P _m ≦ d
a = a2... P _m > d

In this embodiment, two linear functions of the intake pipe pressure P _m representing the engine load factor KLoff at the time of steady state operation is stored in advance in the ROM34 in the form shown in equation (44). In this case, the intake pipe pressure d at the connection point CP is the same as the intake pipe pressure d at the connection point CP for the engine load factor KLon described above. Therefore, the number of parameters can be further reduced. Of course, the intake pipe pressures at these connection points CP may be different from each other.
The parameters a and r in the equation (44) are calculated based on the following equations.
a1 = a1 ^* · ktha
a2 = a2 ^* · ktha
c = c ^* · ktha · kpa
Here, a1 ^* , a2 ^* , and c ^* are the engine load factor at the gradient and the connection point when the engine ambient environment state is the above-described reference environment state, that is, the standard state, respectively.

Therefore, considering that the atmospheric temperature correction coefficient ktha or the atmospheric pressure correction coefficient kpa is a representative value representative of the actual engine ambient environment condition, the reference environment condition is based on the representative value representative of the actual engine environment environment condition. In other words, the parameters a1 ^* , a2 ^* , and c ^* are corrected. Alternatively, it can be considered that the engine load factor KLoff in the reference environment state is corrected based on the representative value representative of the actual engine ambient environment state.

On the other hand, in consideration of the fact that the engine charging efficiency depends on the engine speed NE, in this embodiment, the parameters a ^* (a1 ^* , a2 ^* ), c ^* are set according to the engine speed NE.

More specifically, as shown in FIG. 40A, the gradient a1 ^* increases as the engine speed NE increases when the engine speed NE is low, and the engine speed NE when the engine speed NE is high. It becomes smaller as NE becomes higher. Further, as shown in FIG. 40B, the gradient a2 ^* increases as the engine speed NE increases when the engine speed NE is low, and the engine speed NE increases when the engine speed NE is high. As it gets smaller. These gradients a1 ^* and a2 ^* are obtained in advance by experiments and are stored in advance in the ROM 34 in the form of maps shown in FIGS. 40A and 40B as functions of the engine speed NE.

Further, as shown in FIG. 41, the engine load factor c ^{* at} the connection point CP increases as the engine speed NE increases when the engine speed NE is low, and the engine speed NE when the engine speed NE is high. Becomes smaller as becomes higher. The engine load factor c ^{* at the} connection point CP is also obtained in advance by experiments, and is stored in advance in the ROM 34 in the form of a map shown in FIG. 41 as a function of the engine speed NE.

Therefore, generally speaking, for a plurality of engine speeds NE different from each other, two in-cylinder charged fresh air amount M _c-air at the time of steady operation or intake pipe pressure P _m representing engine load factor KLoff. This means that a linear function formula is obtained and stored in advance.
Then, by detecting the intake pipe pressure P _m by, for example, a pressure sensor 39, equation (43) above from the detected intake pipe pressure P _m or (44) accurately and the engine load factor KLon or KLoff using Thus, the air-fuel ratio can be easily determined, and the air-fuel ratio can be accurately and easily matched with the target air-fuel ratio.

Thus the engine load factor KLon, that the KLoff expressed by a linear function of the intake pipe pressure P _m is the engine load factor KLon, it is not necessary to create a map representing the relationship between the KLoff and the intake pipe pressure P _m Therefore, the effort to create a map is eliminated first. It also means that it is not necessary to solve complicated differential equations, for example, so that the calculation load on the CPU 35 is reduced.

By the way, as can be seen by referring to the equation (25), it can be said that the engine load factor KL represents the in-cylinder charged fresh air amount _Mc-air . Here, when the EGR gas is not supplied, considering that only fresh air is filled into the cylinder, the engine load factor KLoff when the EGR gas is not supplied is filled into the cylinder at this time. the total amount of gas, i.e., can be considered to represent the in-cylinder charged gas amount M _c.

Here, assuming that the in-cylinder charged gas amount _Mc does not change between when the EGR gas is supplied and when the EGR gas is not supplied, the engine load factor KLoff when the EGR gas is not supplied is the EGR gas. There not only in-cylinder charged gas amount M _c when not supplied, it comes to show respectively a cylinder charged gas amount M _c when the EGR gas is supplied.

On the other hand, as described above, the in _- cylinder charged fresh air amount Mc _-air during steady operation and when EGR gas is supplied is represented by the engine load factor KLon.
Therefore, the result ΔKL (= KLoff−KLon) obtained by subtracting the engine load factor KLon when the EGR gas is supplied from the engine load factor KLoff when the EGR gas is not supplied is the in-cylinder filling during the steady operation. This means that the EGR gas amount M _c-egr is represented.

Specifically, for example, as shown in FIG. 42, P _m = if a KLoff = KLoff1 when P _m 1 is and KLon = KLon1 is cylinder charged EGR gas during normal operation The quantity M _c-egr is expressed by ΔKL (= KLoff1-KLon1).

Accordingly, the in-cylinder charged EGR gas amount M _c-egr during steady operation can be calculated based on the following equation (45).
M _c-egr = kegr1 · ΔKL (45)
Here, kegr1 is a conversion coefficient from the engine load factor KL to the in-cylinder charged EGR gas amount Mc _-egr , and KLoff and KLon are engine load factors calculated from the above equations (43) and (44), respectively. Represents.

Therefore, if the intake pipe pressure P _m is calculated by the above-described calculation formula or is detected by the pressure sensor, the cylinder filling EGR gas at the time of steady operation is calculated from the intake pipe pressure P _m by using the above-described formula (45). The quantity M _c-egr can be determined accurately and easily.

Incidentally, as described above, during steady operation, the EGR control valve passage gas flow rate m _egr and the cylinder intake EGR gas amount m _c-egr are equal to each other, and the cylinder filling EGR gas amount M _c-egr is equal to the cylinder intake EGR gas. It is represented by the product of the quantity m _c-egr and ΔT (sec) (M _c-egr = m _c-egr · ΔT).
Therefore, the above-described difference ΔKL also represents the EGR control valve passage gas flow rate m _egr during steady operation.

In the present embodiment, the EGR control valve passage gas flow rate m _egr during steady operation is calculated based on the following equation (46).
m _egr = kegr2 · ΔKL (46)
Here, kegr2 represents a conversion coefficient from the engine load factor KL to the EGR control valve passage gas flow rate m _egr , and KLoff and KLon are engines calculated from the above equations (43) and (44), respectively. It represents the load factor.

As described so far, the EGR control valve passage gas flow rate m _egr at the time of steady operation is calculated using the above equation (46). However, the EGR control valve passage gas flow rate m _egr during transient operation can also be calculated using this equation (46).
That is, the EGR control valve passage gas flow rate m _egr greatly depends on the pressure difference before and after the EGR control valve 22, that is, the difference between the exhaust pressure _Pe and the intake pipe pressure P _m, and is upstream of the EGR control valve 22 during transient operation. given the exhaust pressure P _e and the exhaust temperature T _e is not much different exhaust pressure P _e and the exhaust temperature T _e during steady-state operation, whether transient operation that it would be during the steady operation, the intake pipe pressure P _m If determined, the EGR control valve passage gas flow rate m _egr is determined.

Therefore, the EGR control valve passage gas flow rate m _egr during steady operation and transient operation can be accurately and easily obtained from the intake pipe pressure P _m using the above-described equation (46). In this case, the in-cylinder charged EGR gas amount M _c-egr at the time of steady operation can be calculated from the EGR control valve passage gas flow rate m _egr at the time of steady operation, or from the difference ΔKL using the above-described equation (45). It can also be calculated.

FIG. 43 shows a routine for calculating the EGR control valve passage gas flow rate m _egr in this embodiment described above. This routine is executed by interruption every predetermined time.

Referring to FIG. 43, first, at step 100, the intake pipe pressure P _m , the engine speed NE, and the EGR opening degree θ _e are read. In the subsequent step 101, an atmospheric temperature correction coefficient k _{t a} and an atmospheric pressure correction coefficient k p _a are calculated. In the subsequent step 102, the intake pipe pressure d ^* and the engine load ratios c ^* and r ^* at the connection point CP under the reference environmental condition are calculated from the maps of FIGS. 36, 37 (B), and 41. The In the next step 103, parameters d, c, r are calculated by correcting d ^* , c ^* , r ^* by ktha, kpa. In step 104, the detected intake pipe pressure P _m is whether following the intake pipe pressure d at the connection point is determined. When P _m ≦ d, the routine proceeds to step 105, where a1 ^* and e1 ^* are calculated from the maps of FIG. 35 (C) and FIG. 40 (A). In the subsequent step 106, the gradients a ^* and e ^* are set to a1 ^* and e1 ^* , respectively. Next, the routine proceeds to step 109. On the other hand, when P _m > d, the routine proceeds to step 107, where a2 ^* and e2 ^* are calculated from the maps of FIG. 35 (D) and FIG. 40 (B). In the following step 108, the gradients a ^* and e ^* are set to a2 ^* and e2 ^* , respectively. Next, the routine proceeds to step 109.

In step 109, parameters a and e are calculated by correcting a ^* and e ^* with ktha and kpa. In the subsequent step 110, the engine load factor KLoff is calculated based on the equation (44) (KLoff = a · (P _m −d) + c). In the subsequent step 111, the engine load factor KLon is calculated based on the equation (43) (KLon = e · (P _m −d) + r). In the following step 112, the difference ΔKL is calculated (ΔKL = KLoff−KLon). In the following step 113, the EGR control valve passage gas flow rate m _egr is calculated based on the equation (45) (m _egr = kegr2 · ΔKL).

  In the above-described embodiment, the engine load factors KLoff and KLon are each represented by two linear function equations. However, the engine load factors KLoff and KLon can also be expressed by n m-order function equations (n, m = 1, 2,...), Respectively.

Thus, in the embodiment described above, the first function formula cylinder charging fresh air amount or the engine load factor KLoff when the EGR gas is not being supplied and at the time of steady operation which is a function expression of the intake pipe pressure P _m the first function formula previously obtained is stored a function expression of the cylinder charging fresh air amount or the engine load factor KLon intake pipe pressure P _m when and EGR gas in the time of steady state operation is supplied with representative And the second function equation is obtained and stored in advance, and the in-cylinder filling is performed from the obtained intake pipe pressure P _m by using the first and second function equations. The fresh air amount or engine load factor KLoff, KLon is calculated, the difference ΔKL between the calculated in-cylinder charged fresh air amount or engine load factor KLoff, KLon is calculated, and the EGR control valve passage gas amount m _{egr is set} to the difference ΔKL. Calculation based on And it comes to that.

More generally speaking, the in-cylinder charged fresh air amount or engine load factor KLoff at the time of steady operation and when EGR gas is not supplied, and the in-cylinder at the time of steady operation and when EGR gas is supplied The difference ΔKL from the charged fresh air amount or the engine load factor KLon is expressed by a function expression of the intake pipe pressure P _m , and the function expression is obtained and stored in advance to obtain the intake pipe pressure P _m, and the obtained intake pipe The difference ΔKL is calculated from the pressure P _m using the functional equation, and the EGR control valve passage gas flow rate m _egr during steady operation and transient operation, and the in-cylinder charged EGR gas amount M _c-egr during steady operation, The calculation is based on this difference ΔKL.

Next, another embodiment of the present invention will be described.
The above-described difference ΔKL can be expressed as the following equation (47) using equations (44) and (43) representing KLoff and KLon, respectively.
ΔKL = KLoff−KLon
= (A-e) · ( P m -d) + (c-r) (47)
Here, when (a−e) = h and (c−r) = i are replaced, Equation (47) becomes as follows.
ΔKL = h · (P _m −d) + i (48)
h = h1... P _m ≦ d
h = h2... P _m > d

Therefore, as shown in FIG. 44, the difference ΔKL is expressed by two linear function expressions of the intake pipe pressure P _m having different slopes and continuous at the connection point CP. That is, the difference ΔKL is expressed by a linear function expression of the gradient h1 when the intake pipe pressure P _m is low, and by a linear function expression of the gradient h2 when the intake pipe pressure P _m is high.

In the present embodiment, two linear function expressions of the intake pipe pressure P _m representing the difference ΔKL are stored in advance in the ROM 34 in the form shown in Expression (48). In this way, the number of parameters can be further reduced.
Each parameter h, d, i in the equation (48) is calculated based on the following equation.
h1 = h1 ^* · ktha
h2 = h2 ^* · ktha
i = i ^* · ktha · kpa
Here, h1 ^* , h2 ^* , and i ^* are the differences in the gradient and the connection point CP, respectively, when the engine ambient environment state is the reference environment state. These h1 ^*, h2 ^*, i ^* Figure 45 is obtained by experiment in advance, as a function of the engine speed NE and the EGR opening theta _e respectively (A), (B), of a map shown in (C) Is stored in the ROM 34 in advance. The parameter d is the same as that in the above-described embodiment, and thus the description thereof is omitted.

Therefore, generally speaking, for a plurality of different EGR opening degrees θ _e , two linear function expressions of the intake pipe pressure P _m representing the difference ΔKL are respectively obtained and stored in advance. Also, different with respect to a plurality of engine speed NE, it becomes that the two linear functions of the intake pipe pressure P _m representing the difference ΔKL are stored previously obtained one another.

FIG. 46 shows a calculation routine of the EGR control valve passage gas flow rate m _egr in another embodiment of the present invention described above. This routine is executed by interruption every predetermined time.

Referring to FIG. 46, first, at step 120, the intake pipe pressure P _m , the engine speed NE, and the EGR opening degree θ _e are read. In the following step 121, the atmospheric temperature correction coefficient ktha and the atmospheric pressure correction coefficient kpa are calculated. In the subsequent step 122, the intake pipe pressure d ^* and the difference i ^* at the connection point CP under the reference environmental condition are calculated from the maps of FIGS. 36 and 45C. In the following step 123, parameters d and i are calculated by correcting d ^* and i ^* with ktha and kpa. In step 124, the detected intake pipe pressure P _m is whether following the intake pipe pressure d at the connection point is determined. When P _m ≦ d, the routine proceeds to step 125, where h1 ^* is calculated from the map of FIG. In the following step 126, the gradient h ^* is set to h1 ^* . Next, the routine proceeds to step 129. On the other hand, when P _m > d, the routine proceeds to step 127, where h2 ^* is calculated from the map of FIG. In the following step 128, the gradient h ^* is set to h2 ^* . Next, the routine proceeds to step 129.

In step 129, the parameter h is calculated by correcting h ^* with ktha and kpa. In the following step 130, the difference ΔKL is calculated based on the equation (48) (ΔKL = h · (P _m −d) + i). In the following step 131, the EGR control valve passage gas flow rate m _egr is calculated based on the equation (46) (m _egr = kegr2 · ΔKL).
Here, the EGR opening degree θ _e will be briefly described. As described above, the EGR opening degree is represented by the step number STP of the step motor of the EGR control valve 22. That is, when the step number STP becomes zero, the EGR control valve 22 is closed, and when the step number STP increases, the EGR opening. The degree also increases.

However, in practice, as shown in FIG. 47, even when the step number STP increases from zero, the EGR control valve 22 does not open immediately, and when the step number STP exceeds STP1, the EGR control valve 22 is finally opened. I speak. For this reason, it is necessary to represent the EGR opening degree θ _e by the result of subtracting STP1 from the step number STP (STP−STP1).
Further, the EGR control valve 22 usually because it contains manufacturing errors, there is a possibility that actual EGR opening theta _e for the number of steps STP is deviated from the normal opening degree. Therefore, in the internal combustion engine shown in FIG. 1, a correction coefficient kg for making the actual EGR opening coincide with the normal opening is obtained, and this correction coefficient kg is added to the step number STP.

Therefore, the EGR opening degree θ _e is expressed based on the following equation.
θ _e = STP−STP0 + kg
Here, STP0 is the number of steps at which the EGR control valve 22 starts to open in the drawing tolerance center product. In the present embodiment, the EGR opening degree θ _e calculated in this way is used as an argument of the map.

Incidentally, the in-cylinder charged EGR gas amount M _c-egr at the calculated the EGR control valve passage gas flow rate m _egr or during steady-state operation as described above may be further corrected in consideration of the exhaust temperature T _e.

To describe a case of correcting the EGR control valve passage gas flow rate m _egr, the EGR control valve passage gas flow rate m _egr in this case is represented for example by the following equation.
m _egr = m _egr・ kwu ・ krtd ・ kinc
Here, kwu represents a warm-up correction coefficient, kr _t d represents a retard correction coefficient, and kinc represents an increase correction coefficient.

The warm-up correction coefficient kwu is for correcting the EGR control valve passage gas flow rate m _egr during engine warm-up operation. That is, during the warm-up operation, the exhaust gas temperature _Te is lower than after the completion of the warm-up operation, and the EGR control valve passage gas flow rate m _egr (g / sec) increases accordingly. Since the EGR control valve passage gas flow rate m _egr calculated using the above equation (43), (44) or equation (48) is the EGR control valve passage gas flow rate after completion of the warm-up operation, it is necessary to correct this. There is.

As shown in FIG. 48 (A), the warm-up correction coefficient kwu decreases as the engine coolant temperature THW representing the degree of warm-up increases, and reaches 1.0 when the temperature TWU indicates the completion of warm-up. Retained. The warm-up correction coefficient kwu is stored in advance in the ROM 34 in the form of a map shown in FIG.
On the other hand, the retard correction coefficient krtd is used to correct the EGR control valve passage gas flow rate m _egr when the ignition timing is retarded. That is, the exhaust gas temperature _Te is higher at the time of retard correction than at the time when the retard correction is not performed, and the EGR control valve passage gas flow rate m _egr decreases accordingly.

  As shown in FIG. 48B, the retardation correction coefficient krtd is 1.0 when the retardation amount RTD is zero, and decreases as the retardation amount RTD increases. The retardation correction coefficient krtd is stored in advance in the ROM 34 in the form of a map shown in FIG.

Further, the increase correction coefficient kinc is used to correct the EGR control valve passage gas flow rate m _egr when the fuel injection amount is increased. That is, the exhaust gas temperature _Te is lower during the increase correction than when the increase correction is not performed, and the EGR control valve passage gas flow rate m _egr increases accordingly.
As shown in FIG. 48C, the increase correction coefficient kinc is 1.0 when the increase correction Finc is zero, and increases as the increase correction Finc increases. This increase correction coefficient kinc is stored in advance in the ROM 34 in the form of a map shown in FIG.
In this way, the EGR control valve passage gas flow rate m _egr can be obtained with higher accuracy.

Incidentally, the engine operating state the exhaust temperature T _e when the ignition timing retard correction or fuel increasing correction is not performed (e.g., the engine speed NE and required load L) determined in advance as a function of the actual exhaust temperature T detecting or estimating _e, so as to correct the EGR control valve passage gas flow rate m _egr based on the difference between the actual exhaust temperature T _e and the exhaust temperature T _e when the retard correction or fuel increasing correction is not performed May be. The same is true for the in-cylinder charged EGR gas amount M _c-egr during the steady operation, and a description thereof is omitted.

  In the internal combustion engine shown in FIG. 1, as described above, the EGR passage 21 downstream of the EGR control valve 22 is branched and connected to the intake branch pipe 12 of each cylinder. In this configuration, in order to suppress variation in the amount of EGR gas supplied to each cylinder, as shown in FIG. 49, a throttle 23 is provided in each EGR passage 21 downstream of the EGR control valve 22. Can do.

In this case, first, during steady operation, the throttle passage gas flow rate mchk (g / sec), which is the flow rate of the EGR gas passing through the throttle 23, matches the EGR control valve passage gas flow rate m _egr . Therefore, as can be seen from the above description, the throttle passage gas flow rate mchk during steady operation can be calculated based on the difference ΔKL. The throttle passage gas flow rate mchk represents the flow rate of the EGR gas flowing into the intake pipe portion.

On the other hand, during transient operation, the throttle passage gas flow rate mchk does not necessarily match the EGR control valve passage gas flow rate m _egr . However, when the volume in the EGR passage 21 from the EGR control valve 22 to the throttle 23 is relatively small, mchk substantially matches m _egr . Accordingly, when the volume in the EGR passage 21 from the EGR control valve 22 to the throttle 23 is relatively small, the throttle passage gas flow rate mchk is calculated based on the difference ΔKL, whether in steady operation or transient operation. It will be possible.

In the present embodiment described so far, for example, the engine load factor KLon, KLoff or the difference ΔKL from the intake pipe pressure P _m estimated using the calculation model or the intake pipe pressure P _m detected by the pressure sensor 39. Is calculated. However, for example, the intake pipe pressure P _m is estimated based on the throttle opening or the output of an air flow meter arranged in the intake duct 14 upstream of the throttle valve 17, and the engine load factor is calculated from the estimated intake pipe pressure P _m. KL can also be calculated.
Here, when the intake pipe pressure P _m is estimated based on the throttle opening, the intake pipe pressure P _m is a function of the throttle opening θ _t , the engine speed NE, and the EGR opening θ _e. It is obtained in advance and stored in the form of a map.

On the other hand, when the intake pipe pressure P _m is estimated based on the output of the air flow meter, the estimated intake pipe pressure P _m is the maximum pressure P that can be taken by the intake pipe pressure P _m for the detection accuracy of the air flow meter. _May exceed _mmax . However, in the region where P _m > P _mmax , as shown in FIG. _50B, the engine load factor KLon expressed by the above equation (43) is more than the engine load factor KLoff expressed by the equation (44). In this case, the difference ΔKL becomes a negative value. In other words, the estimated intake pipe pressure P _m is exceeds the maximum pressure P _mmax, there may not be accurately calculating the difference [Delta] KL.

Therefore, as shown in FIG. _50A , in the region where P _m > P _mmax , such a problem can be eliminated if the difference ΔKL is held at a constant value ΔKLC. That is, when the estimated intake pipe pressure P _m exceeds the maximum pressure P _mmax, it is possible to continue to accurately calculate the difference [Delta] KL.

  Incidentally, as shown in FIG. 51, a pair of intake passages 13a and 13b separated from each other by a partition wall 24 are provided in the intake branch pipe 13 corresponding to each cylinder 5, and in one intake passage 13b. An internal combustion engine in which a swirl control valve 25 is arranged is known.

  For example, the swirl control valve 25 is closed during engine low load operation and opened during engine high load operation. As shown in FIG. 51 (A), when the swirl control valve 25 is opened, gas flows into the cylinder 5 from both intake passages 13a and 13b, and therefore a sufficient amount of fresh air enters the cylinder 5. Is supplied. On the other hand, as shown in FIG. 51 (B), when the swirl control valve 25 is closed, the gas flows into the cylinder 5 only from one intake passage 13a. A surrounding swirl flow is formed.

Here, when the swirl control valve 25 is closed, as described above, the gas flows into the cylinder 5 from only one of the intake passages 13a, so the flow path from the EGR control valve 22 to the cylinder 5 Is smaller than the case where the swirl control valve 25 is opened.
Further, when the swirl control valve 25 is closed, the gas flow velocity in the vicinity of the outlet of the EGR passage 21 is increased. In particular, as shown in FIG. 51, when the EGR passage 21 is connected to the intake passage 13a in the region separated by the partition wall 24, the EGR passage 21 when the swirl control valve 25 is closed. The increase in gas flow velocity in the vicinity of the outlet of the gas is large.

Thus, when the swirl control valve 25 is closed, the volume of the flow path from the EGR control valve 22 to the cylinder 5 becomes smaller than when the swirl control valve 25 is opened, and the EGR Since the flow velocity of the gas in the vicinity of the outlet of the passage 21 is increased, the time required for the exhaust gas to reach the cylinder 5 after passing through the EGR control valve 22 is shortened, and the EGR passage 21 downstream of the EGR control valve 22 is used. The diffusion of exhaust gas in the inside and the intake passage 13a is promoted.
That is, this means that when the swirl control valve 25 is closed, the above-described dead time Td2 is shortened and the above-described time constant τ2 is small as compared with the case where the swirl control valve 25 is opened. Become.

Therefore, when the internal combustion engine includes the swirl control valve 25, in the above-described embodiment, the dead time Td2 with the engine speed NE as a variable based on the state where the swirl control valve 25 is opened and A map of the time constant τ2 is obtained in advance by experiments or the like and stored in the ROM 34. When the swirl control valve 25 is opened, a dead time Td2 and a time constant τ2 obtained from this map based on the engine speed NE are obtained. In-cylinder inflow exhaust gas flow rate m _egr-egr is calculated.

The maps of the dead time Td2 and the time constant τ2 here are in the form as shown in FIGS. 17A and 17B, respectively.
On the other hand, when the swirl control valve 25 is closed, the dead time Td2 and the time constant τ2 are obtained from the map shown in FIGS. 17A and 17B based on the engine speed NE, and these dead time Td2 and hour The cylinder inflow exhaust gas flow rate m _egr-egr is calculated using the dead time Td2 calculated by multiplying the constant τ2 by a correction coefficient smaller than 1.0 and the time constant τ2.
According to this, even when the internal combustion engine includes a swirl control valve, the in-cylinder inflow exhaust gas flow rate can be calculated more accurately.

  More generally speaking, the technical idea is that the passage volume from the EGR control valve to the intake valve is changed according to the engine operating state, or the flow passage cross-sectional area of the intake passage is changed according to the engine operating state. The present invention is equally applicable to an internal combustion engine in which the gas flow velocity in the vicinity of the outlet of the EGR passage that opens to the intake passage is changed.

By the way, in the internal combustion engine provided with the swirl control valve as shown in FIG. 51, when the swirl control valve is closed, as described above, the EGR control is performed as compared with the case where the swirl control valve is opened. Since the volume of the flow path from the valve to the cylinder is reduced and the flow velocity of the gas in the vicinity of the outlet of the EGR passage is increased, the amount of EGR gas filled in the cylinder is increased.
That is, when the swirl control valve is closed, the amount of EGR gas charged in the cylinder increases, and conversely, the amount of fresh air charged in the cylinder decreases, and therefore the engine load factor is small. Become.

  Therefore, when the internal combustion engine is provided with the swirl control valve, when calculating the engine load factor KLon when the EGR gas is supplied into the cylinder using the above equation (43), the swirl control valve is It is preferable to consider that the amount of EGR gas filled in the cylinder varies depending on whether the valve is opened or closed.

Therefore, in the case where the internal combustion engine includes the swirl control valve 25, in the above-described embodiment, the parameter e (e1 ^* ,) used in the equation (43) is based on the state in which the swirl control valve 25 is opened. A map for setting e2 ^* ), d ^* , r ^* is obtained by experiment or the like and stored in the ROM 34. When the swirl control valve 25 is opened, each parameter e (e1 ^* , e2 ^* ), d ^* , r ^* are obtained, and the engine load factor KLon is calculated from the equation (43) based on these parameters.

On the other hand, when the swirl control valve 25 is closed, the parameters e (e1 ^* , e2 ^* ), d ^* , r ^* are obtained from the above-mentioned map, and a correction coefficient larger than 1.0 is set in at least one of these parameters. The engine load factor KLon is calculated from the equation (43) based on the parameter calculated by multiplying.
According to this, even when the internal combustion engine includes a swirl control valve, the engine load factor KLon can be calculated more accurately.
More generally speaking, the technical idea is that the passage volume from the EGR control valve to the intake valve is changed according to the engine operating state, or the flow passage cross-sectional area of the intake passage is changed according to the engine operating state. The present invention is equally applicable to an internal combustion engine in which the gas flow velocity in the vicinity of the outlet of the EGR passage that opens to the intake passage is changed.

In addition, when the in-cylinder inflow exhaust gas flow rate is calculated directly instead of the in-cylinder inflow exhaust gas flow rate, the swirl control valve is closed and the in-cylinder inflow exhaust gas flow rate increases. In some cases, the in-cylinder fresh air flow rate may be corrected so as to decrease earlier than when the swirl control valve is opened.
Further, in the above formulas (43) and (44), the parameter d ^* is a value obtained from the same map in formulas (43) and (44) in the above embodiment, but is used in formula (43). The parameter d ^* used and the parameter d ^* used in the equation (44) may be values obtained from different maps.
In addition, since the amount is calculated if time is taken for the flow rate, the flow rate in the above-described embodiment substantially means the amount.

  The present invention is particularly applicable to an internal combustion engine in which the EGR passage is connected to the intake branch pipe (or intake port) so that the exhaust gas supplied to the intake branch pipe via the EGR passage is accumulated there. This is advantageous.

It is a figure which shows the whole internal combustion engine provided with the control apparatus of this invention. It is a figure which shows the cylinder inflow gas amount model applicable to the internal combustion engine which is not equipped with an EGR apparatus. It is a figure which shows the relationship between a throttle opening and a flow coefficient. It is a figure which shows the relationship between throttle opening and opening cross-sectional area. Is a diagram showing the function Φ a (P _m / P _a). It is a figure which shows the basic concept of a throttle model. It is a figure which shows the basic concept of the intake pipe model applicable to the internal combustion engine which is not equipped with an EGR apparatus. It is a figure which shows the basic concept of an intake valve model. It is a figure regarding the definition of the cylinder filling gas amount and the cylinder inflow gas flow rate. It is a figure which shows the cylinder inflow gas amount model of this invention applicable to an internal combustion engine provided with the EGR apparatus. It is a figure which shows the basic concept of the intake pipe model applicable to the internal combustion engine provided with the EGR apparatus. It is a figure which shows the basic concept in calculation of an EGR control valve passage gas flow rate. It is a figure showing the relationship between an engine load factor and exhaust pressure. It is a figure showing the relationship between an EGR control valve passage gas flow rate and exhaust temperature. It is a figure which shows the change of an inflow exhaust gas partial flow rate when an EGR control valve passage gas flow rate changes. It is a figure which shows the dead time Td1 regarding the inflow exhaust gas partial flow rate, and time constant (tau) 1. It is a figure which shows the dead time Td2 regarding the in-cylinder inflow exhaust gas flow rate, and time constant (tau) 2. 7 is a flowchart showing a routine for calculating a cylinder inflow exhaust gas flow rate m _egr-egr . It is a figure which shows dead time Td and time constant (tau). It is a figure which shows the dead time Td1 regarding the inflow exhaust gas partial flow rate, and the number N1 of annealing. It is a figure which shows the dead time Td2 regarding the in-cylinder inflow exhaust gas flow rate, and the number of annealing N2. 7 is a flowchart showing a routine for calculating a cylinder inflow exhaust gas flow rate m _egr-egr . 7 is a flowchart showing a routine for calculating a cylinder inflow exhaust gas flow rate m _egr-egr in consideration of a distribution coefficient for each cylinder. 7 is a flowchart showing a routine for calculating a cylinder inflow exhaust gas flow rate m _egr-egr in consideration of a distribution coefficient for each cylinder, a dead time, and a time constant. It is a figure which shows the relationship between intake valve closing timing CA, dead time Td2, and time constant (tau) 2. It is a figure which shows the relationship between the correction coefficient Ktd regarding the intake valve closing timing CA, the dead time Td2, and the correction coefficient Kτ regarding the time constant τ2. It is a figure which shows the relationship between intake valve closing timing CA, dead time Td2, and time constant (tau) 2. It is a figure which shows the relationship between the correction coefficient Ktd regarding the intake valve closing timing CA, the dead time Td2, and the correction coefficient Kτ regarding the time constant τ2. It is a figure which shows the relationship between intake valve closing timing CA, dead time Td2, and time constant (tau) 2. It is a figure which shows the relationship between the correction coefficient Ktd regarding the intake valve closing timing CA, the dead time Td2, and the correction coefficient Kτ regarding the time constant τ2. It is a figure for _{demonstrating the} EGR control valve passage gas amount m _egr . Exhaust pressure P _e, is a graph showing the exhaust temperature T _e, and P _e / √T _e. It is a diagram which shows a function ( _Pm / _Pe ). Is a diagram showing an example of a relationship between the engine load factor KLon and the intake pipe pressure P _m. It is a diagram which shows gradient e1, e2. It is a diagram which shows the intake pipe pressure d in a connection point. It is a diagram which shows the engine load factor r in a connection point. Is a diagram showing an example of a relationship between the engine load factor KLon and the intake pipe pressure P _m. Is a diagram showing an example of a relationship between the engine load ratio KLoff and the intake pipe pressure P _m. It is a diagram which shows gradient a1, a2. It is a diagram which shows the engine load factor c in a connection point. It is a diagram for demonstrating difference (DELTA) KL. It is a flowchart which shows the calculation routine of the EGR control valve passage gas flow rate m _egr . Is a diagram showing an example of the relationship between the difference ΔKL the intake pipe pressure P _m. It is a diagram which shows the difference i in gradient h1, h2 and a connection point. It is a flowchart which shows the calculation routine of the EGR control valve passage gas flow rate m _egr by another embodiment of this invention. It is a graph showing the relationship between the EGR opening degree theta _e and the number of steps STP. It is a diagram which shows various correction coefficients. FIG. 6 is a partial view of an internal combustion engine showing still another embodiment of the present invention. Intake pipe pressure P _m is a diagram showing the engine load factor KL when exceeding the maximum pressure P _mmax. It is a figure which shows the intake branch pipe of the internal combustion engine provided with the swirl control valve.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Engine main body 5 Combustion chamber 6 Intake valve 7 Intake port 8 Exhaust valve 11 Fuel injection valve 13 Intake pipe 18 Throttle valve 22 EGR control valve

Claims (12)

An internal combustion engine in which an intake gas passage and an exhaust passage are connected via an exhaust gas recirculation passage and an exhaust gas flow rate control valve for controlling the flow rate of exhaust gas flowing in the exhaust gas recirculation passage is disposed in the exhaust gas recirculation passage In the engine, the in-cylinder inflow exhaust gas amount that is the amount of exhaust gas flowing into the cylinder of the internal combustion engine is calculated using the passing exhaust gas amount that is the amount of exhaust gas that passes through the exhaust gas flow control valve In the in-cylinder inflow exhaust gas amount calculation apparatus, the in-cylinder inflow exhaust gas with respect to the dead time corresponding to the time taken for the exhaust gas that has passed through the exhaust gas flow control valve to reach the cylinder and the change in the amount of exhaust gas passing through the cylinder In the in-cylinder inflow exhaust gas amount calculating device that calculates the inflow in-cylinder gas amount in consideration of the follow-up delay of the change in the gas amount,
The follow-up delay is a first-order delay, the time constant of the first-order delay and the dead time are changed according to the engine speed, and the dead time is calculated as the engine speed increases. apparatus. An internal combustion engine in which an intake gas passage and an exhaust passage are connected via an exhaust gas recirculation passage and an exhaust gas flow rate control valve for controlling the flow rate of exhaust gas flowing in the exhaust gas recirculation passage is disposed in the exhaust gas recirculation passage In the engine, a passing exhaust gas amount that is the amount of exhaust gas passing through the exhaust gas flow control valve is calculated using a parameter that changes the passing exhaust gas amount, and the calculated passing exhaust gas amount is used. Then, in the in-cylinder inflow exhaust gas amount calculation device that calculates the in-cylinder inflow exhaust gas amount that is the amount of exhaust gas flowing into the cylinder of the internal combustion engine, the value of the parameter is read, and the read value is used as the exhaust value. Reflects the dead time corresponding to the time it takes for the exhaust gas that has passed through the gas flow control valve to reach the cylinder, and the follow-up delay of the change in the in-cylinder inflow exhaust gas amount with respect to the change in the passing exhaust gas amount. In the cylinder inflow exhaust gas amount calculation device which calculates the passing exhaust gas amount by using the read value in which the dead time and tracking lag are reflected,
The follow-up delay is a first-order delay, the time constant of the first-order delay and the dead time are changed according to the engine speed, and the dead time is calculated as the engine speed increases. apparatus.   An internal combustion engine includes a plurality of cylinders, and a ratio of an actual in-cylinder inflow exhaust gas amount in each cylinder to the calculated in-cylinder inflow exhaust gas amount is obtained in advance as a distribution coefficient, and the calculated cylinder The in-cylinder inflow exhaust gas amount calculation device according to claim 1 or 2, wherein the inflow cylinder exhaust gas amount in each cylinder is calculated by multiplying the inflow exhaust gas amount by the distribution coefficient.   The in-cylinder inflow exhaust gas amount calculation device according to any one of claims 1 to 3, wherein the internal combustion engine includes a plurality of cylinders, and the follow-up delay and the dead time are set for each cylinder.   The internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow rate control valve to the intake valve, or is controlled according to the engine operating state to change the flow passage sectional area of the intake passage. Means for changing the flow rate of the gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage, and the set value of the following delay and the set time of the dead time according to the passage volume or the flow rate of the gas. The in-cylinder inflow exhaust gas amount calculation device according to any one of claims 1 to 4, wherein is changed.   A function equation expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage is obtained in advance and stored, and the function equation is used to store the passing exhaust gas amount in the intake passage. 2. The in-cylinder inflow exhaust gas amount calculation device for an internal combustion engine according to claim 1, wherein the exhaust gas amount passing through the internal combustion engine is controlled in accordance with an engine operating state and the intake valve is controlled from the exhaust gas flow rate control valve. Means for changing the flow volume of the exhaust gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage by changing the cross-sectional area of the intake passage according to the engine operating state A cylinder inflow exhaust gas amount calculation device that changes parameters other than the pressure in the intake passage according to the passage volume or the gas flow rate. An internal combustion engine in which an exhaust gas recirculation passage is connected to the intake passage and the exhaust passage and an exhaust gas flow rate control valve for controlling the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage is disposed in the exhaust gas recirculation passage In an engine, an intake air amount, which is an amount of exhaust gas flowing into the intake passage, is calculated using a passing exhaust gas amount, which is an amount of exhaust gas passing through the exhaust gas flow control valve. In the in-passage exhaust gas amount calculation device, a dead time corresponding to a time taken for the exhaust gas that has passed through the exhaust gas flow control valve to reach the intake passage, and an inflow into the intake passage with respect to the change in the passing exhaust gas amount In the intake passage inflow exhaust gas amount calculation device for calculating the inflow exhaust gas amount in the intake passage in consideration of the follow-up delay of the change in the exhaust gas amount,
The follow-up delay is a primary delay, the time constant of the primary delay and the dead time are changed according to the engine speed, and the dead time is reduced as the engine speed increases. Calculation device. An internal combustion engine in which an exhaust gas recirculation passage is connected to the intake passage and the exhaust passage and an exhaust gas flow rate control valve for controlling the flow rate of the exhaust gas flowing in the exhaust gas recirculation passage is disposed in the exhaust gas recirculation passage In the engine, a passing exhaust gas amount that is the amount of exhaust gas passing through the exhaust gas flow control valve is calculated using a parameter that changes the passing exhaust gas amount, and the calculated passing exhaust gas amount is used. Then, in the intake passage inflow exhaust gas amount calculation device for calculating the inflow exhaust gas amount in the intake passage, which is the amount of exhaust gas flowing into the intake passage, the value of the parameter is read, and the exhaust value The dead time corresponding to the time it takes for the exhaust gas that has passed through the gas flow control valve to reach the intake passage, and the change in the amount of exhaust gas flowing into the intake passage with respect to the change in the amount of exhaust gas that has passed through are added. A delay to reflect, in the dead time and the intake passage inflow exhaust gas amount calculation device which calculates the passing exhaust gas amount by using the read value following delay is reflected,
The follow-up delay is a primary delay, the time constant of the primary delay and the dead time are changed according to the engine speed, and the dead time is reduced as the engine speed increases. Calculation device.   The internal combustion engine has a plurality of cylinders, and the exhaust gas recirculation passage is connected to an intake passage connected to each cylinder, and is connected to each cylinder corresponding to the calculated exhaust gas amount in the intake passage. A ratio of the actual intake passage inflow exhaust gas amount to the intake passage thus obtained is obtained in advance as a distribution coefficient, and the calculated intake air passage exhaust gas amount is multiplied by the distribution coefficient to connect to each cylinder. The intake passage inflow exhaust gas amount calculation device according to claim 7 or 8, wherein the exhaust passage inflow exhaust gas amount into the intake passage is calculated.   The intake passage inflow exhaust gas amount calculation device according to any one of claims 7 to 9, wherein the internal combustion engine includes a plurality of cylinders, and the follow-up delay and the dead time are set for each cylinder. .   The internal combustion engine is controlled according to the engine operating state to change the passage volume from the exhaust gas flow rate control valve to the intake valve, or is controlled according to the engine operating state to change the flow passage sectional area of the intake passage. Means for changing the flow rate of the gas in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage, and the set value of the following delay and the set time of the dead time according to the passage volume or the flow rate of the gas. The apparatus for calculating an inflow exhaust gas amount in an intake passage according to any one of claims 7 to 10, wherein is changed.   A function equation expressed as a function of the pressure in the intake passage and a parameter other than the pressure in the intake passage is obtained in advance and stored, and the function equation is used to store the passing exhaust gas amount in the intake passage. 8. An apparatus for calculating an exhaust gas amount flowing into an intake passage according to claim 7, wherein the internal combustion engine is controlled in accordance with an engine operating state to calculate the exhaust gas amount passing through from the pressure of the exhaust gas flow control valve to the intake valve. A means for changing the passage volume, or a gas flow velocity in the vicinity of the opening of the exhaust gas recirculation passage that opens to the intake passage by changing the cross-sectional area of the intake passage controlled according to the engine operating state. An intake passage inflow exhaust gas amount calculation device comprising means for changing a parameter other than the pressure in the intake passage according to the passage volume or the gas flow rate.

Images