FUEL SUPPLY METHOD

Abstract

The method of the present invention calculates based on the status of the exhaust emission purifier, a required supply of fuel to be supplied from the fuel supply valve to the exhaust passage, reads a unit fuel supply LU in accordance with a energized time tU for the fuel supply valve per one shot, supplies with a driving cycle depending on the required supply, fuel in the unit supply to the exhaust passage, when EDT−EDU<EA−(LU−g) is established, the energized time tUT corresponding to the target fuel supply is interpolated as a function of (LU/g) and the target fuel supply LU is updated as a new unit supply LU, and the function of (LU/g) is supplied as a new energized time tU from the fuel supply valve to the exhaust passage.

Description

TECHNICAL FIELD

The present invention relates to a fuel supply method for allowing an internal-combustion engine including an exhaust emission purifier to efficiently perform the activation processing and the regenerating processing of the exhaust emission purifier.

BACKGROUND ART

In recent years, for complying with strict exhaust gas regulations applied to an internal combustion engine, it is necessary to activate an oxidation catalytic converter which comprises an exhaust emission purifier at the time of starting up the engine, or to maintain an activated status of the converter in an actuation of the engine. Therefore, Patent Literature 1 has proposed an internal combustion engine in which an exhaust gas heating system is incorporated in an exhaust passage upstream of the exhaust emission purifier. The exhaust gas heating system generates a burning gas in an exhaust gas, and supplies the generated burning gas to the exhaust emission purifier at the downstream thereof, thereby activating the purifier and maintaining the activated status thereof. Therefore, the exhaust gas heating system is generally provided with a fuel supplying valve for supplying fuel to the exhaust passage and an igniting unit such as a glow plug for heating the fuel to be ignited, thereby generating a burning gas.

CITATION LIST

Patent Literature

• PTL: Japanese Patent Laid-open No. 2010-059836

SUMMARY OF INVENTION

Technical Problem

When controlling the fuel supplied from a fuel supply valve to an exhaust passage, the hunting of the control is preferably avoided by minimizing a change in the exhaust temperature or a change in the air/fuel ratio caused by the fuel supplied from the fuel supply valve to the exhaust passage.

On the other hand, the fuel supply valve for supplying fuel to the exhaust passage in the conventional exhaust heating unit disclosed in PTL 1 for example basically has the same configuration as that of a fuel injection valve for injecting fuel pressurized at a predetermined driving cycle. Thus, the hunting of the control can be effectively avoided by minimizing the driving cycle to the fuel supply valve, by minimizing the energized time per one shot, and by minimizing the amount of fuel supplied to the exhaust passage during the energized time per one shot.

However, in the case of the conventional fuel supply valve, it has been known that the reduction of the energized time per one shot causes a sudden increase of the variation of the amount of the fuel supplied to the exhaust passage. This is caused by the structure of the fuel supply valve itself and an influence by the viscosity of the fuel itself, meaning that different amounts of fuel are supplied to the exhaust passage depending on the manufacturing tolerances of the individual fuel supply valves. Due to the technical background of the fuel supply valve as described above, blindedly minimizing the energized time per one shot is essentially impossible. Thus, the energized time per one shot has been conventionally determined so that the maximum variation error of the fuel supply caused by the manufacturing tolerances of the individual fuel supply valves for example is equal to or lower than the maximum tolerance of the amount of the fuel supplied to the exhaust passage during the energized time per one shot to the fuel supply valve.

Thus, the conventional exhaust heating unit has a disadvantage as described below. That is, when the amount of fuel supplied to the exhaust passage is not so high, the driving cycle of the fuel supply valve is increased to thereby deteriorate the fuel ignitability. Furthermore, a change in the exhaust temperature or the air/fuel ratio is increased depending on when fuel is supplied and when no fuel is supplied, thus causing the hunting phenomenon of the control.

It is an objective of the present invention to provide a method for supplying fuel that can reduce the energized time per one shot to the fuel supply valve than in the case of the conventional structure.

Solution to Problem

The present invention is in a method for supplying fuel from a fuel supply valve to an exhaust passage at an upstream side of an exhaust emission purifier, the method comprises the steps of calculating, based on the status of the exhaust emission purifier, a required supply of fuel to be supplied from the fuel supply valve to the exhaust passage; reading a unit fuel supply L_U to be supplied to the exhaust passage in accordance with a energized time t_U to the fuel supply valve per one shot; intermittently supplying, with a driving cycle depending on the required supply, fuel of the unit fuel supply L_U from the fuel supply valve to the exhaust passage; reading the maximum tolerance E_A corresponding to the unit supply L_U; reading the maximum variation error E_DU of the fuel supply valve corresponding to the unit supply L_U; calculating, regard to the fuel the unit supply L_U, an actual fuel supply g actual supplied to the exhaust passage; setting a target fuel supply L_UT that is less than the unit supply L_U by a certain amount; reading the maximum variation error E_DT of the fuel supply valve corresponding to the target fuel supply L_UT; judging whether E_DT−E_DU<E_A−(L_U−g) is established or not; interpolating, when it is judged that E_DT−E_DU<E_A−(L_U−g) is established, a energized time T_UT to the fuel supply valve corresponding to the target fuel supply L_UT as a function of (L_U/g); and updating the target fuel supply L_UT as a new unit fuel supply L_U and using the function of (L_U/g) as a new energized time t_U to drive the fuel supply valve to supply fuel to the exhaust passage.

In the method for supplying fuel according to the present invention, the step of reading a unit fuel supply L_U to be supplied to the exhaust passage in accordance with a energized time t_U to the fuel supply valve per one shot may read the latest updated unit supply L_U.

When fuel to be supplied from the fuel supply valve in accordance with the required supply at every driving cycle of the fuel supply valve is in an amount exceeding the double of the unit fuel supply, a half of the to-be-supplied amount may be supplied.

The state of the exhaust emission purifier for calculating the required supply is a temperature of the exhaust emission purifier or an air/fuel ratio of exhaust flowing therein, and the method may further comprise a step of judging, by carrying-out of the step of intermittently supplying fuel from the fuel supply valve to the exhaust passage for the energized time t_U, whether the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent or not. In this instance, only when it is judged that the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent, the step of judging whether E_DT−E_DU<E_A−(L_U−g) is established or not may be carried out. In this case, the step of judging whether the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent or not may include a step of judging whether at least one of the temperature of the exhaust emission purifier and the change rate thereof is within a predetermined range or not, or whether at least one of the air/fuel ratio of the exhaust flowing in the exhaust emission purifier and the change rate thereof is within a predetermined range or not.

When the sum of a detection error of the air/fuel ratio and a detection error of an air-intake amount is less than the maximum tolerance, the step of judging whether E_DT−E_DU<E_A−(L_U−g) is established or not may be carried out.

The method may further comprise a step of judging whether an amount of EC passing through the exhaust emission purifier has a value equal to or less than a predetermined value. In this instance, when it is judged that an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value, the step of judging whether E_DT−E_DU<E_A−(L_U−g) is established or not may be carried out.

The method may further comprise a step of judging whether a value E_T/ΔT_C obtained by dividing a detection temperature error E_T of the exhaust emission purifier by the rate ΔT_C of temperature increase of the exhaust emission purifier per unit time smaller than the maximum tolerance E_A or not. In this instance, when it is judged that E_T/ΔT_C<E_A is established, the step of judging whether E_DT−E_DU<E_A−(L_U−g) is established or not may be carried out. In this case, the method may further comprise a step of judging whether an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value. In this instance, only when it is judged that an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value, the step of judging whether E_T/ΔT_C<E_A is established or not may be carried out.

Advantageous Effects of Invention

According to the method for supplying fuel of the present invention, a unit supply of fuel can be reduced without exceeding the maximum tolerance. This can consequently reduce the driving cycle of the fuel supply valve, thus suppressing the hunting phenomenon of the control than in the case of the conventional structure.

In order to read the unit supply L_U of fuel supplied to the exhaust passage, the updated latest unit supply L_U can be read to thereby improve the initial accuracy when the fuel supply is reduced.

when the fuel amount to be supplied from the fuel supply valve at every driving cycle of the fuel supply valve in accordance with a required supply exceeds the double of the unit fuel supply, a half of the to-be-supplied amount can be supplied to thereby suppress a sudden change in the fuel supply per one shot. This can consequently further suppress the hunting phenomenon of the control due to a change in the fuel supply. Furthermore, the main effect of the invention can be continuously obtained by continuously supplying the unit supply of fuel as long as possible.

Only when the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent, a step is performed to determine whether E_DT−E_DU<E_A−(L_U−g) is established or not. By doing this, a control under a transitional control can be avoided. This can consequently further suppress the hunting phenomenon of the control. In particular, a control under a transitional control can be securely avoided by judging whether at least one of the temperature of the exhaust emission purifier and the change rate thereof is within a predetermined range or whether at least one of the air/fuel ratio of the exhaust flowing in the exhaust emission purifier and the change rate thereof is within a predetermined range or not.

When the sum of the detection error of the air/fuel ratio and the detection error of the air-intake amount is lower than the maximum tolerance, whether E_DT−E_DU<E_A−(L_U−g) is established or not can be judged to thereby securely maintain a reliable control.

When the HC amount passing through the exhaust emission purifier is judged to be equal to or less than a predetermined value, whether E_DT−E_DU<E_A−(L_U−g) is established or not can be judged to thereby securely maintain a reliable control.

When it is judged that the value E_T/ΔT_C obtained by dividing the detection temperature error E_T of the exhaust emission purifier by the rate ΔT_C of temperature increase of the exhaust emission purifier per unit time is lower than the maximum tolerance E_A, whether E_DT−E_DU<E_A−(L_U−g) is established or not can be judged to thereby obtain a similar effect. In particular, only when the HC amount passing through the exhaust emission purifier is judged to be equal to or less than the predetermined value, whether E_T/ΔT_C<E_A or not can be judged to thereby further maintain a reliable control.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 is a conceptual diagram schematically illustrating a vehicle engine system including an exhaust heating unit according to the present invention;

FIG. 2 is a control block diagram illustrating the main part of the embodiment shown in FIG. 1;

FIG. 3 is a graph schematically illustrating the required supply corresponding to the driving cycle of the furl supply in the embodiment shown in FIG. 1;

FIG. 4 is a graph schematically illustrating the energized time of the fuel supply valve and the maximum tolerance thereof and the maximum variation error of the individual fuel supply valves;

FIG. 5 is a graph schematically illustrating the characteristic of an airflow meter;

FIG. 6 is a graph schematically illustrating the characteristic of an air/fuel ratio sensor;

FIG. 7 is a graph schematically illustrating the relation between the amount of HC in the exhaust and the lean shift amount of the detection value by the air/fuel ratio sensor;

FIG. 8 is a map schematically illustrating the relation between the reaction rate of the catalyst temperature and the catalyst and the HC amount in the exhaust and the O₂ concentration;

FIG. 9 is a flowchart showing, together with FIG. 10, the procedure for setting, in the embodiment shown in FIG. 1, the fuel supply from the fuel supply valve per one shot in a catalyst activation mode;

FIG. 10 is a flowchart showing, together with FIG. 9, the procedure for setting the fuel supply from the fuel supply valve per one shot in the catalyst activation mode;

FIG. 11 is a flowchart showing the details of the subroutine of the fuel addition in the flowchart shown in FIG. 9;

FIG. 12 is a flowchart showing, together with FIG. 13, the procedure for setting the fuel supply from the fuel supply valve per one shot in a catalyst regenerating mode; and

FIG. 13 is a flowchart showing, together with FIG. 12, the procedure for setting the fuel supply from the fuel supply valve per one shot in the catalyst regenerating mode.

DESCRIPTION OF EMBODIMENTS

An embodiment in which the present invention is applied to a compression ignition type internal combustion engine will be in detail explained with reference to FIG. 1 to FIG. 13. The present invention is not, however, limited to the embodiment, and the construction thereof may be freely modified corresponding to required characteristics. The present invention is effectively applied to a spark ignition type internal combustion engine in which gasoline, alcohol, LNG (Liquefied Natural Gas) or the like is used as fuel to be ignited by a spark plug, for example.

FIG. 1 schematically illustrates the main part of the engine system in this embodiment. FIG. 2 illustrates the control blocks of the main part. FIG. 1 does not illustrate a valve actuating mechanism for the intake and exhaust of the engine 10 and a throttle mechanism and a silencer as well as a general EGR apparatus as an auxiliary machine of the engine 10 for example. FIG. 1 also does not illustrate a part of various sensors required for the smooth operation of the engine 10 and the above-described auxiliary machines.

The engine 10 in this embodiment is a multicylinder internal-combustion engine in which fuel of light oil is directly injected from a fuel injection valve 11 to a combustion chamber 10a in a compressed state to thereby cause spontaneous ignition. However, according to the characteristic of the present invention, the engine 10 also may be a single-cylinder internal-combustion engine.

A cylinder head 12 includes an intake port 12a and an exhaust port 12b opposed to the combustion chamber 10a. The cylinder head 12 includes a valve actuating mechanism (not shown) including an intake valve 13a for opening and closing the intake port 12a and an exhaust valve 13b for opening and closing the exhaust port 12b. The fuel injection valve 11 opposed to the center of the upper end of the combustion chamber 10a is also mounted on the cylinder head 12 so as to be positioned between the intake valve 13a and the exhaust valve 13b.

The amount and the injection timing of the fuel injected from the fuel injection valve 11 into the combustion chamber 10a is controlled by the Electronic Control Unit (ECU) 15 based on the vehicle operating condition including the depression amount of the accelerator pedal 14 by the driver. The depression amount of the accelerator pedal 14 is detected by an accelerator opening sensor 16. The detection information is outputted to the ECU 15.

The ECUI 15 includes: an operating status determining section 15a for determining the vehicle operating condition based on the information from this accelerator opening sensor 16 and various sensors for example (which will be described later); a fuel injection setting unit 15b; and a fuel injection valve driving unit 15c. The fuel injection setting unit 15b sets, based on the determination result by the operating status determining section 15a, the injection amount and the injection timing of fuel from the fuel injection valve 11. The fuel injection valve driving unit 15c controls the operation of fuel injection valve 11 so that the fuel in an amount set by the fuel injection setting unit 15b is injected from the fuel injection valve 11 at the set timing.

The cylinder block in which the piston 17a reciprocates is mounted on a crank angle sensor 18. The crank angle sensor 18 detects the rotation phase of the crankshaft 17c connected via a connecting rod 17b to the piston 17a (i.e., a crank angle) to output this to the ECU 15. The operating status determining section 15a of the ECU 15 determines, based on the information from this crank angle sensor 18, the rotation phase of the crankshaft 17c and the engine rotational speed as well as the vehicle speed for example on a real-time basis.

An air-intake pipe 19 connected to the cylinder head 12 so as to communicate with the intake port 12a defines the air-intake passage 19a together with the intake port 12a. The exhaust pipe 20 connected to the cylinder head 12 so as to communicate with the exhaust port 12b defines the exhaust passage 20a together with the exhaust port 12b.

An exhaust turbo-supercharger (hereinafter simply referred to as a supercharger) 21 is provided so as to connect the air-intake pipe 19 and the exhaust pipe 20. This supercharger 21 uses the kinetic energy of the exhaust flowing in the exhaust passage 20a to supercharge the combustion chamber 11a to increase the intake filling efficiency. The supercharger 21 in this embodiment is a turbocharger in which the main part is composed of a compressor 21a and an exhaust turbine 21b rotating with this compressor 21a in an integrated manner. The compressor 21a is provided on the midway of the air-intake pipe 19 positioned at the upstream of the surge tank 19b provided on the midway of the air-intake pipe 20. The exhaust turbine 21b is provided on the midway of the exhaust pipe 20 connected to the cylinder head 12 so as to communicate with the exhaust port 12b. In order to reduce the intake temperature heated via the compressor 21a by the heat transfer from the exhaust turbine 21b subjected to a high-temperature exhaust, an intercooler 21c is provided. This intercooler 21c is provided on the midway of the air-intake passage 19a between the compressor 21a and the surge tank 19b provided on the midway of the air-intake pipe 19.

The air-intake pipe 19 provided at the upstream of the compressor 21a of the supercharger 21 includes an airflow meter 22 that detects the flow rate V_A of the intake flowing in the air-intake passage 19a (which will be hereinafter referred to as an air-intake amount) to output this to the ECU 15.

An exhaust emission purifier 23 is provided on the midway of the exhaust pipe 20 between the exhaust turbine 21b of the supercharger 21 and a silencer (not shown). The exhaust emission purifier 23 functions to detoxify the toxic substance generated by the combustion of mixed air in the combustion chamber 10a. The exhaust emission purifier 23 in this embodiment includes, in an order from the upstream side, generally well-known Nitrogen Oxides (NO_X) storage catalytic converter 23a, Diesel Particulate Filter (DPF) 23b, and an oxidation catalytic converter 23c.

An exhaust heating unit 24 is provided on the midway of the exhaust passage 20a between the exhaust port 12b and the exhaust turbine 21b of the supercharger 21. This exhaust heating unit 24 functions to heat the exhaust from the engine 10 to the exhaust emission purifier 23 to activate the oxidation catalytic converter 23c of the exhaust emission purifier 23 and to maintain the active condition or functions to subject the DPF 23b to a regenerating processing or the NO_X storage catalytic converter 23a to a deoxidation processing. The exhaust heating unit 24 in this embodiment includes a fuel supply valve 24a and a glow plug 24.

The amount of the fuel supplied from the fuel supply valve 24a mounted on the exhaust pipe 20 to the exhaust passage 20a is set, based on the determination result by a fuel supply requirement determining section 15d of the ECU 15 (which will be described later), by the fuel supply setting section 15e of the ECU 15. The fuel supply valve driving section 15f of the ECU 15 controls the operation of the fuel supply valve 24a so that the fuel in an amount set by fuel supply setting section 15e is supplied from the fuel supply valve 24a to the exhaust passage 20a.

The glow plug 24b for igniting the fuel supplied from the fuel supply valve 24a to the exhaust passage 20a is fixed to the exhaust pipe 20 so that the heat-generating portion thereof protrudes to the exhaust passage 20a to be opposed to an injection region of the fuel injected from the fuel supply valve 24a. This glow plug 24b is connected to an in-vehicle power source (not shown) via the glow plug driving section 15g of the ECU 15. The glow plug driving section 15g switches ON/OFF of the conduction of the glow plug 24b based on the determination result by the fuel supply requirement determining section 15d of the ECU 15.

The exhaust passage 20a at the upstream side of the fuel supply valve 24a includes a first exhaust temperature sensor 25. This first exhaust temperature sensor 25 detects the exhaust temperature T_I flowing in the DPF 23b to output this to the ECU 15.

The exhaust passage 20a between the NO_X storage catalytic converter 23a and the DPF 23b has an air/fuel ratio sensor 26. This air/fuel ratio sensor 26 detects the air/fuel ratio R_N of the exhaust flowing therein to output this to the ECU 15. The DPF 23b and the oxidation catalytic converter 23d have therebetween a catalyst temperature sensor 27 that detects the temperature T_C of the oxidation catalytic converter 23c to output this to the ECU 15. The exhaust passage 20a at the downstream side of the oxidation catalytic converter 23c has the second exhaust temperature sensor 28 that detects the exhaust temperature T_O having passed through the oxidation catalytic converter 23c to output this to the ECU 15.

The fuel supply requirement determining section 15d of the ECU 15 determines, based on the determination result of the operation condition by the operating status determining section 15a, the necessity of the activation of the oxidation catalytic converter 23c and the necessities of the regenerating processing in the DPF 23b and the NO_X deoxidation processing in the NO_X storage catalytic converter 23a. When this fuel supply requirement determining section 15d determines that the activation of the oxidation catalytic converter 23c as well as the regenerating processing in the DPF 23b and the NO_X deoxidation processing in the NO_X storage catalytic converter 23a are required, then the addition of fuel through the fuel supply valve 24a will be performed. The determination result by this fuel supply requirement determining section 15d is outputted to the glow plug driving section 15g of the ECU 15, the fuel supply setting section 15e, and a unit fuel supply updating section 15h (which will be described later).

The fuel supply requirement determining section 15d of the ECU 15 determines that there is a fuel supply requirement (i.e., the exhaust heating unit 24 must be operated) when any of the following cases “a” to “d” occurs.

a: a case where the oxidation catalytic converter 23c is inactive or is expected to be inactive.
b: a case where the DPF 23b is clogged by the deposition of HC.
a case where the storage of NO_X by the NO_X storage catalytic converter 23a is in a saturated condition.
d. a case where the regenerating processing of the DPF 23b is required even when the DPF 23b is not clogged.
The case “a” can be judged based on the temperature information T_I, T_O, and T_C from the first exhaust temperature sensor 25 and the second exhaust temperature sensor 26 as well as the catalyst temperature sensor 27. The case “b” can be judged based on the accumulated operating time of the engine 10 or the accumulated fuel injection amount from the fuel injection valve 11 for example and also can be judged by an exhaust pressure sensor. The case “c” can be similarly judged based on the accumulated operating time of the engine 10 or the accumulated fuel injection amount from the fuel injection valve 11 for example.

In order to maintain the correct performance of the exhaust emission purifier 23, the fuel supply setting section 15e of the ECU 15 sets the fuel amount to be supplied to the exhaust passage 20a (hereinafter referred to as a required supply) to L_TA and L_TR.

More specifically, based on the exhaust temperature T_I and the air suction amount V_A per a predetermined time (e.g., 1 second), the activation required fuel supply L_TA to be supplied to the exhaust passage 20a in order to maintain the active condition of the oxidation catalytic converter 23c is set based on the following formula (1). The exhaust temperature T_I is the exhaust temperature that flows at the upstream side of the oxidation catalytic converter 23c and that flows in the exhaust passage 20a proximal to the oxidation catalytic converter 23c. The exhaust temperature T_I will be hereinafter referred to as a catalyst upstream exhaust temperature. The air suction amount V_A per a predetermined time (which will be hereinafter referred to as an air-intake amount) is acquired from the airflow meter 22.

L_TA={(T_L−T_I)V_A·C}/J  (1)

In the formula, T_L means the lowest temperature at which the oxidation catalytic converter 23c is in an active condition and is stored in the fuel supply setting section 15e in advance. C shows the air specific heat. J shows the heat generation amount of the fuel supplied to the exhaust passage 20a and is also stored in the fuel supply setting section 15e in advance. The catalyst upstream exhaust temperature T_I is acquired from the first exhaust temperature sensor 25.

When the fuel supply requirement determining section 15d determines that the regenerating processing by the DPF 23b constituting the exhaust emission purifier or the NO_X deoxidation processing by the NO_X storage catalystic converter is required, then the regenerating required fuel supply L_TR to be supplied to the exhaust passage 20a is set based on the following formula (2).

L_TR=V_A/R_T)−q  (2)

In this formula, R_T shows the air/fuel ratio as a command of the exhaust flowing through the exhaust passage 20a to the exhaust emission purifier 23 (hereinafter referred to as a target air/fuel ratio) and is stored in the fuel supply setting section 15e in advance. In this formula, q shows the injection amount of the fuel injected from the fuel injection valve 11 to the combustion chamber 10a of the engine 10 and is acquired from the fuel injection valve driving unit 15c.

The information regarding the required fuel supplies L_TA and L_TR set by the fuel supply setting section 15e (which will be hereinafter collectively referred to as a required fuel supply L_T for convenience) is outputted to the fuel supply valve driving section 15f and the unit fuel supply updating section 15h of the ECU 15.

The fuel supply valve driving section 15f performs, at every calculation cycle t_P, a processing to multiply the required fuel supply L_T set by the fuel supply setting section 15e with the calculation cycle t_P (e.g., 20 milliseconds) to totalize the resultant values. When the fuel totalized value obtained at every calculation cycle t_P reaches the unit fuel supply L_U updated by the unit fuel supply updating section 15h, the energized time t_U corresponding to this unit fuel supply L_U is given to the fuel supply valve 24a. At the same time, the fuel totalized value is updated to a value obtained by (totalized value−unit supply L_U). Then, the fuel totalization processing is repeated to drive the fuel supply valve 24 intermittently. Therefore, the higher the required fuel supply L_T per a predetermined time is, the shorter the driving interval of the fuel supply valve 24a is. The lower the required supply L_T per a predetermined time is, the longer the driving interval of the fuel supply valve 24a is. The driving information from the fuel supply valve driving 15f to the fuel supply valve 24a is outputted to the surplus tolerance calculating section 15i of the ECU 15.

When the engine 10 is in a transitional operating condition (e.g., in a sudden acceleration requiring the supply of a large amount of fuel within a short time), the driving cycle t_C of the fuel supply valve 24a must be synchronized with the fuel burst interval of the respective cylinders of the engine 10. As a result, there may be a case where, even when the unit fuel supply L_U is supplied to the exhaust passage 20a at every driving cycle t_C matching the fuel burst interval of the respective cylinders of the engine 10, the fuel supply ratio is insufficient to thereby fail to provide an appropriate control. In this embodiment, only when the required supply ΔL_T to be supplied at every driving cycle t_C of the fuel supply valve 24a (=ΔL_A·t_C, ΔL_TR·t_C) exceeds the double of the unit fuel supply L_U, a half of the required fuel supply L_T multiplied with the driving cycle t_C is supplied from the fuel supply valve 24a to the exhaust passage 20a. FIG. 3 schematically illustrates the change of the required fuel supply ΔL_T to be supplied at the respective times t₁ to t₅ with the shortest driving cycle t_C of the fuel supply valve 24a. In FIG. 3, the required fuel supply Δ_T accumulates in the period of time t₁ to time t₃ to exceed the unit fuel supply L_U. At the time t₃ and time t₄ at which the required fuel supply L_T at every driving cycle t_C exceeds the double of the unit fuel supply L_U, not the unit fuel supply L_U but fuel in an amount of ΔL_T/2 is supplied. Thus, after the time t₅, the required fuel supply ΔL_T at every driving cycle t_C is convergent to a value smaller than the double of the unit fuel supply L_U. Thus, the unit fuel supply L_U is supplied as in the case of the time t₁ and time t₂. This can consequently suppress a delay of a control and can suppress a sudden change of the fuel supply supplied from the fuel supply valve 24a to the exhaust passage 20a.

The ECU 15 includes a convergence determining section 15j in addition to the above-described operating status determining section 15a and fuel supply valve driving section 15f for example.

The convergence determining section 15j determines, based on the determination result of the operating condition from the operating status determining section 15a, whether the oxidation catalytic converter 23c reaches, by the supply of fuel from the fuel supply valve 24a, to the target activation temperature T_T and is in a stable condition or not. The convergence determining section 15j also determines whether the air/fuel ratio R_N reaches, by the supply of fuel from the fuel supply valve 24a, to the target air/fuel ratio R_T and is in a stable condition or not.

The convergence determining section 15j determines that the temperature of the oxidation catalytic converter 23c is convergent in the vicinity of the target activation temperature T_T and is in a stable condition when the following two conditions are satisfied. The first condition is that that the absolute value of the value obtained by deducting from the target activation temperature T_T the temperature of the exhaust flowing in the exhaust passage 20a proximal to the oxidation catalytic converter 23c at the downstream side of the oxidation catalytic converter 23c (which will be hereinafter referred to as a catalyst downstream exhaust temperature) is lower than a positive threshold value T_R set in advance. The second condition is that the absolute value of a rate dT_O of change in the exhaust temperature T_O is smaller than the threshold value dT_R set in advance (which is generally has a positive value close to 0). As a result, it can be securely judged that the engine 10 does not have a transitional operating condition. However, another configuration also may be used where the oxidation catalytic converter 23c has a temperature convergent to the neighborhood of the target activation temperature T_T only when any of the conditions is satisfied.

Similarly, the convergence determining section 15j determines that the exhaust flowing in the exhaust passage 20a has the air/fuel ratio R_N that is convergent ant stable at the neighborhood of the target air/fuel ratio R_T when the following two conditions are satisfied. The first condition is that the absolute value of the value obtained by deducting, from the target air/fuel ratio R_T, the air/fuel ratio R_N acquired by the air/fuel ratio sensor 26 is smaller than the positive threshold value dR_R set in advance. The second condition is that the rate dR_N of change of the air/fuel ratio R_N has an absolute value smaller than the positive threshold value dR_R set in advance (which is generally has a positive value close to 0). As a result, it can be securely judged that the engine 10 does not have a transitional operating condition. However, another configuration also may be used where the exhaust flowing in the exhaust passage 20a has the air/fuel ratio R_N that is convergent in the vicinity of the target air/fuel ratio R_T only when any of the conditions is satisfied.

By the determination processing by the convergence determining section 15j as described above, it is possible to confirm that the engine 10 is not in a transitional operating condition. Thus, the exhaust heating unit 24 can be subjected to a smooth control. This determination result is outputted to the surplus tolerance calculating section 15j.

FIG. 4 schematically illustrates the relation among the unit fuel supply L_U and the maximum tolerance E_A as well as the maximum variation error E_D. It is noted that values are not positive or negative values around 0 as a center but are conveniently represented as absolute values on the basis of percentage. The maximum tolerance E_A shows, with regard to the unit fuel supply L_U set by the unit fuel supply updating section 15h, a target control (e.g., a dislocation value for which a temperature increase control of the oxidation catalytic converter 23c or the air/fuel ratio control of the exhaust can be convergent). The maximum tolerance E_A can be represented as a value equal to or less than the unit supply L_U. Thus, the maximum tolerance E_A represented by percentage can be basically represented as a fixed value regardless of the magnitude of the unit supply L_U. On the other hand, the maximum variation error E_D of the fuel supply, which is caused by the mechanical characteristic of the fuel supply valve 24a itself or the viscosity of the fuel itself for example, tends to rapidly increase as shown by the solid line of FIG. 4 with the decrease of the unit fuel supply L_U. Thus, when the unit fuel supply L_U is tried to be less than the unit fuel supply at which the maximum tolerance E_A is equal to the maximum variation error E_D (this unit fuel supply will be hereinafter referred to as a reference unit fuel supply L_UC), the maximum variation error E_D must be considered. Theerfore, the maximum tolerance E_A must be further estimated to be smaller in consideration of the difference between the maximum variation error E_D and the maximum tolerance E_A so that the unit fuel supply L_U has an error that is smaller than (E_A−E_D).

The surplus tolerance calculating section 15i stores therein the map as show in FIG. 4. The surplus tolerance calculating section 15i deducts, from the totalized value ΣL_D of the unit fuel supply L_U set by the unit fuel supply updating section 15h, the fuel supply to be actually supplied to correspond to the totalized value ΣL_U (hereinafter referred to as an actual fuel supply) to calculate the supply error E_U. Then, the surplus tolerance calculating section 15i deducts the calculated supply error E_U from the maximum tolerance E_A corresponding to the current unit supply L_U stored in the map of FIG. 4 to calculate the amount ΔE_A of the surplus tolerance. The totalized value ΣL_U is an instruction value of the amount of the fuel that is driven, during the detection period of the intake amount V_A, by an instruction from the fuel supply valve driving section 15f and that is supplied from the fuel supply valve 24a to the exhaust passage 20a. Thus, the supply error E_U can be represented by (Σ_U−g). The actual fuel supply g is calculated based on the following formula.

g=(ΔT_I·V_A·C)  (3)

In the formula, ΔT_I shows a difference between T_I detected by the first exhaust temperature sensor 25 and the exhaust temperature T_O detected by the second exhaust temperature sensor 28 and is represented by ΔT_I=T_O−T_I. Thus, the amount ΔE_A of the surplus tolerance is represented as shown by the following formula (4).

ΔE_A=E_A−ΣL_U+(ΔT_I·V_A·C/J)  (4)

The amount ΔE_A of the surplus tolerance calculated by the surplus tolerance calculating section 15i is outputted to the unit fuel supply updating section 15h.

The unit fuel supply updating section 15h in this embodiment updates the fuel supply per one energized time t_U supplied from the fuel supply valve 24a to the exhaust passage 20a (i.e., the unit fuel supply L_U) and outputs the updated unit fuel supply L_U to the fuel supply valve driving section 15f. In this embodiment, the initial value L_US of the unit fuel supply L_U is stored in advance in the unit fuel supply updating section 15h. Thus, such a control is carried out that stepwisedly reduces the unit fuel supply L_U from this initial value L_US by a certain amount of the target unit fuel supply L_UT. Basically, the unit fuel supply L_U is updated to have a smaller value so that, with regard to the required fuel supply L_T set by the fuel supply setting section 15e, the unit fuel supply L_U is supplied with the shortest addition interval (the driving cycle t_C of the fuel supply valve 24a) as possible. In this case, the initial value L_US of the unit fuel supply is set to a value sufficiently higher than the reference unit fuel supply L_UC. The target unit fuel supply L_UT is set to a value smaller than the current unit fuel supply L_U by a predetermined amount and is updated to have a smaller value, when possible, by the update processing.

More particularly, fuel of the target unit fuel supply L_UT smaller than the current unit fuel supply L_U by a certain amount is set. A different processing is performed depending on the case where the target unit fuel supply L_UT is higher than the reference unit fuel supply L_UC and the case where the target unit fuel supply L_UT is lower than the reference unit fuel supply L_UC. Specifically, when the target unit fuel supply L_UT is higher than the reference unit fuel supply L_UC, there is no need to consider the maximum variation error E_D. Thus, when the supply error E_U calculated by the surplus tolerance calculating section 15i is lower than the maximum tolerance E_A, the target unit fuel supply L_UT is updates as a new unit fuel supply L_U.

When the target unit fuel supply L_UT is lower than the reference unit fuel supply L_UC, then the target unit fuel supply L_UT is set to have a value smaller than the current unit fuel supply L_U by a certain amount. The maximum variation error E_DT of the target unit fuel supply L_UT and the maximum variation error E_DU of the current unit fuel supply L_U are read out from the map of FIG. 4. Then, the amount ΔE_D of the surplus variation error obtained by deducting from the maximum variation error E_DT corresponding to the target unit fuel supply L_UT the maximum variation error E_DU corresponding to the current unit fuel supply L_U (=E_DT−E_DU) is compared with the amount ΔE_A of the surplus tolerance given from the surplus tolerance calculating section 15i. Then, when it is judged that the amount ΔE_A of the surplus tolerance is lower than the amount ΔE_D of the surplus variation error, the target unit fuel supply L_UT is updated as a new unit fuel supply L_U.

In any of the cases, the energized time t_U corresponding to the updated unit fuel supply L_U is interpolated as shown in the following formula (5).

t_U=t_UT>k(ΣL_U/g)  (5)

The energized time t_U corresponding to the updated unit fuel supply L_U also can be interpolated by any appropriate function formula other than the formula (5) including (ΣL_U/g) as a variable. For example, the following formula (6) also can be used.

t_U=t_UT+(ΣL_U/g)  (6)

Furthermore, it is also effective to multiply (ΣL_U/a) in order to suppress a sudden change of the right sides of the formulae (5) and (6), with such an interpolation coefficient that is higher than 0 and that is equal to 0 and that is equal to or lower than 1. By including such an interpolation coefficient in the formulae (5) and (6), it is possible to avoid an adverse influence by an error of sensors for example used to calculate the unit fuel supply L_U and the actual fuel supply g or the calculation of the energized time t_U during a transitional operating condition of the vehicle. This interpolation coefficient can be set in advance based on the magnitude of the error based on the resolution of the sensors for example or the vehicle transitional operating condition for example.

As described above, based on the actual fuel supply g corresponding to the unit fuel supply L_U, the energized time t_UT corresponding to the target unit fuel supply L_UT is interpolated. By doing this, even when the current unit fuel supply L_U is reduced to the target unit fuel supply L_UT, the resultant error can be equal to or lower than the maximum tolerance E_A. Thus, when the amount ΔE_A of the surplus tolerance exceeds the amount ΔE_D of the surplus variation error and the current unit fuel supply L_U is reduced to the target unit fuel supply L_UT this exceeds the maximum tolerance E_A. Thus, no update of the unit fuel supply L_U is performed.

As described above, the smaller the unit fuel supply L_U is when compared with the required fuel supply L_T, fuel is supplied from the fuel supply valve 24a to the exhaust passage 20a with a shorter driving cycle t_C. As a result, fuel supplied to the exhaust passage 20a shows a smaller temporal fluctuation range, thus providing a smaller fluctuation range of the air/fuel ratio R_N in the exhaust in particular.

The unit fuel supply updating section 15h in this embodiment includes an update availability determining section 151 to determine the availability of the above-described update processing of the unit fuel supply L_U and the energized time t_U. When this update availability determining section 151 determines that the following conditions (A) (C) (E), and (B) or (D) are satisfied, this update availability determining section 151 allows the update of the unit fuel supply L_U and the energized time t_U by the unit fuel supply updating section 15h. When this update availability determining section 151 determines that the following conditions (A), (C), (E), and (B) or (D) are not satisfied on the other hand, this update availability determining section 151 does not allow the update of the unit fuel supply L_U and the energized time t_U by the unit fuel supply updating section 15h. Then, unit fuel supply updating section 15h stores the current unit fuel supply L_U and energized time t_U as the latest unit fuel supply L_U and energized time t_U.

ΔE_A>ΔE_D  (A)

E_A>E_VA+E_RT  (B)

E_A>(E_T/E_TC)  (C)

E_A>E_U  (E)

With regard to (A), when the amount ΔE_A of the surplus tolerance exceeds the amount ΔE_D of the surplus variation error as described above, the unit fuel supply L_U is not updated because the current unit fuel supply L_U reduced to the target unit fuel supply L_UT in this case exceeds the maximum tolerance E_AT.

With regard to (B), it has been well-known that the airflow meter 22 and the air/fuel ratio sensor 26 have a unique measurement error due to the detection system thereof. FIG. 5 shows the relation between the air-intake amount by the airflow meter 22 and the measurement error. FIG. 6 shows the relation between the air/fuel ratio R_N by the air/fuel ratio sensor 26 and the measurement error. When the regenerating required fuel supply L_TR is calculated based on the above formula (2), the air-intake amount V_A detected by the airflow meter 22 and the air/fuel ratio R_N of the exhaust detected by the air/fuel ratio sensor 26 include the measurement errors E_VA and E_RT as shown in FIG. 5 and FIG. 6, respectively. As a result, some combination of the measurement errors E_VA and E_RT may exceed the maximum tolerance E_A for the regenerating required fuel supply L_TR. To solve this, in this embodiment, the air/fuel ratio R_N of the exhaust is controlled so that, when the sum (%) of the measurement errors E_VA and E_RT of by the airflow meter 22 and the air/fuel ratio sensor 26 exceeds the maximum tolerance E_A, the above target unit fuel supply L_UT is not updated as a new unit fuel supply L_U. The above formula (5) is not also calculated.

With regard to (C), it has been known that the detection value R_N of the air/fuel ratio sensor 26 is dislocated to the lean side in proportion to the HC amount included in the exhaust (so-called lean shift). Thus, the error of the detection value R_N by the air/fuel ratio sensor due to this lean shift amount ΔE_S must be lower than the maximum tolerance E_A of the unit fuel supply L_U. Thus, when the lean shift amount ΔE_S is too large, the above target unit fuel supply L_UT is not updated as a new unit fuel supply L_U. Similarly, the calculation of the above formula (5) also must not be performed. In this embodiment, when the value ω·g obtained by multiplying the HC purification rate ω by the exhaust emission purifier 23 with the actual fuel supply g is lower than the lean shift amount ΔE_S, then the above target unit fuel supply L_UT is updated as a new unit fuel supply L_U and the above formula (5) is calculated. When the value ω·g obtained by multiplying the HC purification rate ω by the exhaust emission purifier 23 with the actual fuel supply g is equal to or larger than the lean shift amount ΔE_S on the other hand, the above target unit fuel supply L_UT is not updated as a new unit fuel supply L_U and the above formula (5) is not performed.

FIG. 7 schematically illustrates the relation between the HC amount included in the exhaust passing through the exhaust passage 20a including the air/fuel ratio sensor 26 and the lean shift amount ΔE_S in the air/fuel ratio sensor 26. The operating status determining section 15a of the ECU 15 stores therein the map as shown in FIG. 7 in advance. The HC amount included in the exhaust is calculated by the fuel supply valve update availability determining section 151 based on the air-intake amount V_A and, the fuel injection amount from the fuel injection valve 11, and the fuel supply from the fuel supply valve 24a. The HC purification rate u by the exhaust emission purifier 23 can be calculated by dividing the HC reaction rate v by the exhaust emission purifier 23 by the exhaust flow rate (which is the air-intake amount V_A in this case). The reaction rate v in the exhaust emission purifier 23 can be calculated based on the relation among the HC amount and O₂ concentration in the exhaust and the catalyst temperature. FIG. 8 illustrates the relation among the HC amount and O₂ concentration in the exhaust and the catalyst temperature as described above. The fuel supply valve update availability determining section 151 stores therein the map as shown in FIG. 8. The fuel supply valve update availability determining section 151 reads the reaction rate V based on the catalyst temperature T_C and the HC amount and O₂ concentration in the exhaust. Next, the air-intake amount V_A by the airflow meter 22 is divided by information to calculate the HC purification rate ω. Then, the HC purification rate ω is multiplied with the actual fuel supply g calculated based on the above formula (3) and the result is compared with the above lean shift amount ΔE_S. Then, only when ΔE_S>(v/V_A)·g is established, the unit fuel supply updating section 15h updates the unit fuel supply L_U and the energized time t_U.

With regard to (D), when a control is carried out based on the activation required fuel supply L_TA, an influence by the detection error of the catalyst temperature sensor 27 must be avoided. Thus, it is judged whether the value (E_T/ΔT_C) obtained by dividing the detection temperature error of the exhaust emission purifier 23 (i.e., the detection error E_T of the catalyst temperature T_C detected by the catalyst temperature sensor 27) by the rate ΔT_C of temperature increase of the exhaust emission purifier 23 per unit time is smaller than the maximum tolerance E_A or not. When E_A>(E_T/ΔT_C) is established. Then, the unit supply L_U and the energized time t_U are updated.

With regard to (E), when the target unit fuel supply L_UT is more than the reference unit fuel supply L_UC and the supply error E_U is larger than the maximum tolerance E_A, the target unit fuel supply L_UT cannot be updated as a new unit fuel supply L_U.

FIG. 9 to FIG. 11 illustrate the flow of the fuel supply control in this embodiment in the catalyst activation mode to maintain the oxidation catalytic converter 23c as described above in an activated state. First, the step S11 determines whether there is a fuel supply request or not. When it is judged that there is a fuel supply request (i.e., when it is judged that the activation of the oxidation catalytic converter 23c in the exhaust emission purifier 23 is required), the processing proceeds to the step S12 to calculate the activation required fuel supply L_TA. Next, the step S13 acquires the unit fuel supply L_U from the unit fuel supply updating section 15h of the ECU 15. Then, the step S14 drives the fuel supply valve 24a to start supplying fuel to the exhaust passage 20a.

FIG. 11 illustrates the subroutine of the fuel supply. Specifically, the step S141 determines whether the required fuel supply ΔL_TA per unit time is less than double of the unit fuel supply L_U or not. When the required fuel supply ΔL_TA per unit time is less than double of the unit fuel supply L_U (i.e., there is no particular problem when fuel is continuously supplied in the unit fuel supply L_U), the processing proceeds to the step S142 to supply fuel with the unit fuel supply L_U. Next, the step S143 determines whether a fuel supply flag is set or not. At an initial stage, no fuel supply flag is set. Thus, the processing proceeds to the step S144 to set a fuel supply flag. Then, the step S145 determines whether there is a fuel supply request or not. When it is judged that there is a fuel supply request (i.e., there is still a need to activate the oxidation catalytic converter 23_C in the exhaust emission purifier 23), then the processing returns to the main flow of FIG. 10 to carry out the step S15 (which will be described later). When the step S145 determines that there is no fuel supply request (i.e., there is no need to activate the oxidation catalytic converter 23_C in the exhaust emission purifier 23), the processing returns to the main flow of FIG. 9 to carry out the step S30 (which will be described later).

On the other hand, when the step S141 determines that the required fuel supply ΔL_TA per unit time exceeds the double of the unit fuel supply L_U (i.e., when it is judged that the continued supply of fuel in the unit fuel supply L_U causes an insufficient fuel supply), the processing proceeds to the step S146. Then, fuel in the amount of a half of the required fuel supply ΔL_TA per unit time is supplied through the fuel supply valve 24a to the exhaust passage 20a. Then, steps after the above step S143 are carried out.

The step S15 determines whether the new target unit fuel supply L_UT set to the current unit fuel supply L_U is less than the reference unit fuel supply L_UC or not. At an initial stage, the newly-set target unit fuel supply L_UT is equal to or more than the reference unit fuel supply L_UC. Thus, the processing proceeds to the step S16 to allow the surplus tolerance calculating section 15i to calculate the supply error E_U. Next, the step S17 determines whether the supply error E_U is smaller than the maximum tolerance E_A or not. When it is judged that the supply error E_U is smaller than the maximum tolerance E_A (i.e., when it is judged that the unit fuel supply L_U can be updated), the processing proceeds to the step S18. Then, the set target unit fuel supply L_UT is updated as a new unit fuel supply L_U and the energized time t_U to the corresponding fuel supply valve 24a is interpolated as shown in the formula (5). Then, the processing again returns to the step S11. When the step S17 determines that the supply error E_U is equal to or larger than the maximum tolerance E_A (i.e., it is judged that the unit fuel supply L_U cannot be updated), the current unit fuel supply L_U and the energized time t_U are directly maintained. Then, the processing again returns to the step S11.

When the above step S15 determines that the target unit fuel supply L_UT newly set to the current unit fuel supply L_U is less than the reference unit fuel supply L_UC, (i.e., when it is judged that the maximum variation error E_D must be considered), then the processing proceeds to the step S19. Then, it is judged whether the absolute value of the value obtained by deducting the current catalyst downstream exhaust temperature T_O from the target activation temperature T_T of the oxidation catalytic converter 23c is smaller than the threshold value T_R or not. When it is judged that the absolute value of the value obtained by deducting the current catalyst downstream exhaust temperature T_O from the target activation temperature T_T of the oxidation catalytic converter 23c is less than the threshold value T_R (i.e., when it is judged that the oxidation catalytic converter 23_C reaches the activation temperature), then the processing proceeds to the step S20. Then, it is judged whether the rate dT_O of exhaust temperature change in downstream of the catalyst is less than the threshold value dT_R or not. When the rate dT_O of exhaust temperature change in downstream of the catalyst is smaller than the threshold value dT_R (i.e., when the oxidation catalytic converter 23_C has a temperature that is converged and stable at the activation temperature), then the processing proceeds to the step S21. Then, the supply error E_U is calculated. The step S22 calculates the amount ΔE_A of the surplus tolerance. The step S23 calculates the amount ΔE_D of the surplus variation error.

When the above step S19 determines that the absolute value of the value obtained by deducting the current catalyst downstream exhaust temperature T_O from the target activation temperature T_T is equal to or larger than the threshold value T_R (i.e., when it is judged that the oxidation catalytic converter 23c is converged at the activation temperature), then the processing returns to the step S11. in this case, it is noted that the current unit fuel supply L_U and energized time t_U are maintained continuously. Similarly, when the step S20 determines that the exhaust temperature change rate dT_CO is equal to or more than the threshold value dT_R (i.e., when the oxidation catalytic converter 23c is not converged at the activation temperature), the processing returns to the step S11 while maintaining the current unit fuel supply L_U and energized time t_U.

After the step S23, the step S24 determines whether the amount ΔE_A of the surplus tolerance is more than the amount ΔE_D of the surplus variation error or not. When it is judged that the amount ΔE_A of the surplus tolerance is larger than the amount ΔE_D of the surplus variation error (i.e., it is judged that the unit fuel supply L_U can be updated to a reduced amount), the processing proceeds to the step S29. Then, it is judged whether the value (E_T/ΔT_U) obtained by dividing the detection error E_T of the catalyst temperature T_C by the rate ΔT_C of temperature increase of the exhaust emission purifier 23 per unit time is less than the maximum tolerance E_A or not. When it is judged that the value of (E_T/ΔT_C) is smaller than the maximum tolerance E_A (i.e., when it is judged that the unit fuel supply L_U can be updated to a reduced amount), the processing proceeds to the step S18. When it is judged that the value of (E_T/T_C) is equal to or larger than the maximum tolerance E_A (i.e., when it is judged that the unit fuel supply L_U cannot be updated to a reduced amount), the current unit fuel supply L_U and energized time t_U are directly maintained and the processing returns to the step S11. Similarly, when the step S24 determines that amount ΔE_A of the surplus tolerance is equal to or less than the amount ΔE_D of the surplus variation error (i.e., when it is judged that an updated unit fuel supply L_U deviates from the maximum tolerance E_A), then the current unit fuel supply L_U and energized time t_U are maintained. Then, the processing returns to the step S11.

On the other hand, when the above step S11 determines that there is no fuel supply request (i.e., it is judged that there is no need to activate the oxidation catalytic converter 23 in the exhaust emission purifier 23), then the processing proceeds to the step S30 to determine whether a fuel supply flag is set or not. When it is judged that the fuel supply flag is set, (i.e., when it is judged that the fuel supply from the fuel supply valve 24a to the exhaust passage 20a continued), then the processing proceeds to the step S31 to stop the fuel supply processing. Then, the step S32 resets the fuel supply flag to thereby complete a series of controls. When the above step S30 determines that no fuel supply flag is set (i.e., when it is judged that the processing to supply fuel from the fuel supply valve 24a to the exhaust passage 20a is not performed), the processing is completed without doing anything.

As described above, when the amount ΔE of the surplus tolerance is less than the amount ΔE_D of the surplus variation error, the unit fuel supply L_U is updated to a smaller target unit fuel supply L_UT. As a result, the oxidation catalytic converter 23c can have a narrower temperature amplitude to reduce the wasteful fuel supply, thus improving fuel consumption. At the same time, even the reduced unit fuel supply L_U can be always less than the maximum tolerance, thus securely maintaining a desired control accuracy.

In the above-described embodiment, the temperature of the oxidation catalytic converter 23c was controlled. However, a similar control configuration also can be used when the air/fuel ratio of the exhaust is controlled.

FIG. 12 and FIG. 13 show the flow of the fuel supply control in this embodiment in the catalyst regenerating mode for performing the regenerating processing by the DPF 23b and the deoxidation processing by the NO_X storage catalytic converter 23a constituting the exhaust emission purifier 23 as described above. The steps S41 to S48, S51 to S54, and S60 to S62 in this embodiment are basically the same as the steps S11 to S18, S21 to S24, and S30 to S32 in the above flowcharts shown in FIG. 9 and FIG. 10. However, the regenerating required fuel supply L_TR in the step S42 is calculated by the formula (2). The fuel supply subroutine of S44 has the same procedure as that of the above embodiment shown in FIG. 11.

When the step S45 determines that the target unit fuel supply L_UT newly set to the current unit fuel supply L_U is less than the reference unit supply L_UC (i.e., when it is judged that the maximum variation error E_D must be considered), the processing proceeds to the step S49. Then, it is judged whether the absolute value of the value obtained by deducting the air/fuel ratio value R_N from the target air/fuel ratio value R_T is smaller than the positive threshold value R_R set in advance or not. When it is judged that the absolute value of the value obtained by deducting the air/fuel ratio value R_N from the target air/fuel ratio value R_T is less than the positive threshold value R_R set in advance (i.e., when it is judged that the current air/fuel ratio R_N substantially reaches the target air/fuel ratio R_T). Then, the processing proceeds to the step S50. Then, it is judged whether the rate dR_N of change of air/fuel ratio detected by the air/fuel ratio sensor 26 has an absolute value that is less than the threshold value dR_R or not. When it is judged that the rate dR_N of change of the air/fuel ratio R_N has an absolute value smaller than the threshold value dR_R (i.e., when it is judged that the exhaust flowing in the exhaust passage 20a has the air/fuel ratio R_N that is converged and stable at the target air/fuel ratio R_T), the processing proceeds to the step S51 to calculate the amount ΔE_A of the surplus tolerance.

When the step S49 determines that the absolute value of the value obtained by deducting the current air/fuel ratio value R_N from the target air/fuel ratio value R_T is equal to or larger than the threshold value R_R(i.e., when it is judged that the current air/fuel ratio R_N is not converged at the target air/fuel ratio R_T), then the processing returns to the step S41. In this case, it is noted that the current unit fuel supply L_U and the energized time t_U are maintained. When the step S50 determines that the absolute value of the rate dR_N of change of the air/fuel ratio is equal to or more than the threshold value R_R (i.e., when it is judged that the exhaust flowing in the exhaust passage 20a has the air/fuel ratio R_N that does not reach the target air/fuel ratio R_T and is unstable), the processing also returns to the step S41. In this case, the current unit fuel supply L_U and the energized time t_U are similarly maintained.

After the step S53, the step S54 determines whether amount ΔE_A of the surplus tolerance is more than the amount ΔE_D of the surplus variation error or not. When the amount ΔE_A of the surplus tolerance is larger than the amount ΔE_D of the surplus variation error, (i.e., when it is judged that the unit fuel supply L_U can be updated and reduced), then the processing proceeds to the step S55. Then, it is judged whether the sum of the measurement errors E_VA and E_RT of the airflow meter 22 and the air/fuel ratio sensor 26 is less than the maximum tolerance E_A or not. When it is judged that the sum of the measurement errors E_VA and E_RT of the airflow meter 22 and the air/fuel ratio sensor 26 is smaller than the maximum tolerance E_A, (i.e., when it is judged that the unit fuel supply L_U can be updated and reduced), the processing proceeds to the step S56. Then, the HC amount in the exhaust is calculated. The step S57 calculates the lean shift amount ΔE_S in the air/fuel ratio sensor 26. The step S58 calculates the catalyst purification rate ω. When the step S55 determines that the sum of the measurement errors E_VA and E_RT is less than the maximum tolerance E_A, (i.e., when it is judged that the unit fuel supply L_U cannot be updated and reduced), then the processing returns to the step S41 while maintaining the current unit fuel supply L_U and energized time t_U.

After the step S58, the step S59 determines whether the above lean shift amount ΔE_S is less than a value obtained by multiplying the HC purification rate ω with the actual fuel supply g or not. When it is judged that the above lean shift amount ΔE_S is less than a value obtained by multiplying the HC purification rate ω with the actual fuel supply g, the processing proceeds to the step S48. When it is judged that the above lean shift amount ΔE_S is equal to or larger than a value obtained by multiplying the HC purification rate ω with the actual fuel supply g (i.e., when it is judged that the unit fuel supply L_U cannot be updated and reduced), then the processing returns to the step S41 while maintaining the current unit fuel supply L_U and energized time t_U.

As described above, similarly in this embodiment, when the amount ΔE of the surplus tolerance is less than the amount ΔE_D of the surplus variation error, the unit fuel supply L_U is updated to a smaller target unit fuel supply L_UT. As a result, the oxidation catalytic converter 23c can have a narrower temperature amplitude to reduce the wasteful fuel supply, thus improving fuel consumption. At the same time, even the reduced unit fuel supply L_U can be always less than the maximum tolerance, thus securely maintaining a desired control accuracy.

It should be noted that, the present invention should be interpreted based only upon the matters described in claims, and in the aforementioned embodiments, all changes and modifications included within the spirit of the present invention can be made other than the described matters. That is, all the matters in the described embodiments are made not to limit the present invention, but can be arbitrarily changed according to the application, the object and the like, including every construction having no direct relation to the present invention.

REFERENCE SIGNS LIST

• • dT_O Rate of exhaust temperature change
  • dT_R Threshold value
  • dR_N Rate of change of air/fuel ratio
  • dR_R Threshold value
  • ΔE_A Surplus amount of tolerance
  • ΔE_D Surplus amount of variation error
  • ΔE_S Lean shift amount in air/fuel ratio sensor
  • ΔL_T Required supply to be supplied at every driving cycle of fuel supply valve
  • ΔT_C Rate of temperature increase of exhaust emission purifier per unit time
  • E_A Maximum tolerance in unit fuel supply
  • E_D Maximum variation error in unit fuel supply
  • E_U Supply error
  • E_RT Measurement error by air/fuel ratio sensor
  • E_T Measurement error by catalyst temperature sensor
  • E_VA Measurement error by airflow meter
  • g Actual fuel supply
  • L_TA Activation required fuel supply
  • L_TR Regenerating required fuel supply
  • L_T Required fuel supply
  • L_U Unit fuel supply
  • L_UC Reference unit fuel supply
  • L_US Initial value of unit fuel supply
  • L_UT Target unit fuel supply
  • R_N Air/Fuel ratio in exhaust
  • R_T Target air/fuel ratio
  • R_R Threshold value
  • t_C Driving cycle of fuel supply valve
  • t₁ to t₅ Time
  • T_C Temperature of oxidation catalytic converter
  • T_O Exhaust temperature at downstream of catalystff
  • T_T Target activation temperature
  • T_R Threshold value
  • v HC Reaction rate in exhaust emission purifier
  • V_A Air-intake amount
  • ω HC Purification rate by exhaust emission purifier
  • 10 Engine
  • 10a Combustion chamber
  • 11 Fuel injection valve
  • 12 Cylinder head
  • 12a Intake port
  • 12b Exhaust port
  • 13a intake valve
  • 13b Exhaust valve
  • 14 Accelerator pedal
  • 15 ECU
  • 15a Operating status determining section
  • 15b Fuel injection setting section
  • 15c Fuel injection valve driving section
  • 15d Fuel supply requirement determining section
  • 15e Fuel supply setting section
  • 15f Fuel supply valve driving section
  • 15g Glow plug driving section
  • 15h Unit fuel supply updating section
  • 15i Surplus tolerance calculating section
  • 15j Convergence determining section
  • 15l Update availability determining section
  • 16 Accelerator opening sensor
  • 17 Cylinder block
  • 17a Piston
  • 17b Connecting rod
  • 17c Crankshaft
  • 18 Crank angle sensor
  • 19 Air-intake pipe
  • 19a Air-intake passage
  • 19b Surge tank
  • 20 Exhaust pipe
  • 20a Exhaust passage
  • 21 Supercharger (Turbocharger)
  • 21a Compressor
  • 21b Exhaust turbine
  • 21c intercooler
  • 22 Airflow meter
  • 23 Exhaust emission purifier
  • 23a NO_X Storage catalytic converter
  • 23b DPF
  • 23c Oxidation catalytic converter
  • 24 Exhaust heating unit
  • 24a Fuel supply valve
  • 24b Glow plug
  • 25 First exhaust temperature detecting sensor
  • 26 Air/Fuel ratio detecting sonsor
  • 27 Catalytic converter temperature detecting sensor
  • 28 Second exhaust temperature detecting sensor

Claims

1-9. (canceled)

10. A method for supplying fuel from a fuel supply valve to an exhaust passage at an upstream side of an exhaust emission purifier, the method comprising the steps of:

calculating, based on the status of the exhaust emission purifier, a required supply of fuel to be supplied from the fuel supply valve to the exhaust passage;

reading a unit fuel supply LU to be supplied to the exhaust passage in accordance with a energized time tU to the fuel supply valve per one shot;

intermittently supplying, with a driving cycle depending on the required supply, fuel of the unit fuel supply LU from the fuel supply valve to the exhaust passage;

reading the maximum tolerance EA corresponding to the unit supply LU;

reading the maximum variation error EDU of the fuel supply valve corresponding to the unit supply LU;

calculating, regard to the fuel the unit supply LU, an actual fuel supply g actually supplied to the exhaust passage;

setting a target fuel supply LUT that is less than the unit supply LU by a certain amount;

reading the maximum variation error EDT of the fuel supply valve corresponding to the target fuel supply LUT;

judging whether EDT−EDU<EA−(LU−g) is established or not;

interpolating, when it is judged that EDT−EDU<EA−(LU−g) is established, an energized time TUT to the fuel supply valve corresponding to the target fuel supply LUT as a function of (LU/g); and

updating the target fuel supply LUT as a new unit fuel supply LU and using the function of (LU/g) as a new energized time tU to drive the fuel supply valve to supply fuel to the exhaust passage.

11. The method as claimed in claim 10, wherein: the step of reading a unit fuel supply LU to be supplied to the exhaust passage in accordance with an energized time tU to the fuel supply valve per one shot reads the latest updated unit supply LU.

12. The method as claimed in claim 10, wherein when fuel to be supplied from the fuel supply valve in accordance with the required supply at every driving cycle of the fuel supply valve is in an amount exceeding the double of the unit fuel supply, a half of the to-be-supplied amount is supplied.

13. The method as claimed in claim 11, wherein when fuel to be supplied from the fuel supply valve in accordance with the required supply at every driving cycle of the fuel supply valve is in an amount exceeding the double of the unit fuel supply, a half of the to-be-supplied amount is supplied.

14. The method as claimed in claim 10, wherein the state of the exhaust emission purifier for calculating the required supply is a temperature of the exhaust emission purifier or an air/fuel ratio of exhaust flowing therein, and

the method further comprises a step of judging, by carrying-out of the step of intermittently supplying fuel from the fuel supply valve to the exhaust passage for the energized time tU, whether the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent or not and, only when it is judged that the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent, the step of judging whether EDT−EDU<EA−(LU−g) is established or not is carried out.

15. The method as claimed in claim 14, wherein the step of judging whether the temperature of the exhaust emission purifier or the air/fuel ratio of the exhaust flowing therein is convergent or not, includes a step of judging whether at least one of the temperature of the exhaust emission purifier and the change rate thereof is within a predetermined range or not, or whether at least one of the air/fuel ratio of the exhaust flowing in the exhaust emission purifier and the change rate thereof is within a predetermined range or not.

16. The method as claimed in claim 10, wherein when the sum of a detection error of the air/fuel ratio and a detection error of an air-intake amount is less than the maximum tolerance, the step of judging whether EDT−EDU<EA−(LU−g) is established or not is carried out.

17. The method as claimed in claim 10, further comprising a step of judging whether an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value, and

when it is judged that an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value, the step of judging whether EDT−EDU<EA−(LU−g) is established or not is carried out.

18. The method as claimed in claim 10, further comprising a step of judging whether a value ET/ΔTC obtained by dividing a detection temperature error ET of the exhaust emission purifier by the rate ΔTC of temperature increase of the exhaust emission purifier per unit time is smaller than the maximum tolerance EA or not, and

when it is judged that ET/ΔTC<EA is established, the step of judging whether EDT−EDU<EA−(LU−g) is established or not is carried out.

19. The method as claimed in claim 18, further comprising a step of judging whether an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value and, only when it is judged that an amount of HC passing through the exhaust emission purifier has a value equal to or less than a predetermined value, the step of judging whether ET/ΔTC<EA is established or not is carried out.