JPH01247753A - Exhaust gas recirculation device - Google Patents

Abstract

PURPOSE:To conduct the control of EGR properly at all times in spite of the production error of an EGR device and the change of time lapse or the like, by correcting through learning at the time of need the valve opening drive force of an EGR control valve which becomes the standard of the EGR(exhaust gas recirculation) control. CONSTITUTION:Part of exhaust gas at an engine E is extracted to a pipe 22 is EGR gas, and is introduced to an intake pipe 9 via a pipe 13 after its flow having been controlled by means of a valve 3. And the valve 3 is driven by means of an electromagnetic valve 23 provided at a pipe 12 which takes out negative pressure, and at the same time, the electromagnetic valve 23 is controlled by means of an ECU4. In this case, respective detection signals from an O2 sensor 14 and the like which detect the quantity of EGR gas, are inputted into the ECU4. And when the state of EGR not being conducted has continued for a fixed time, the valve 3 is gradually driven, and at the same time, the starting time of EGR is detected by means of the O2 sensor 14. As a result, the data of the valve opening starting drive force of the valve 3 is corrected through learning.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to an exhaust gas recirculation device used in an internal combustion engine such as an automobile. This is to learn and correct the driving force for opening the control valve.

<Prior art> One of the effective prevention devices for automobile exhaust gas pollution is an exhaust gas recirculation device (hereinafter referred to as EGR).
There is a t Gas Recirculation device).

This EGR device is a device for reducing nitrogen oxides (NOx), which are the most difficult to purify among the harmful exhaust gases from automobiles. NOx is generated in the combustion chamber of an engine, and the amount of NOx generated increases as the combustion of the air-fuel mixture approaches complete combustion and the combustion temperature increases.

Therefore, the EGR device achieves the above-mentioned purpose by introducing a portion of the exhaust gas, which is an inert gas (hereinafter referred to as EGR gas), into the combustion chamber along with fresh intake air to lower the combustion temperature.

The NOx generation rate decreases as the mixing rate of EGR gas into intake air (hereinafter referred to as EGR rate) increases. However, as a matter of course, at the same time, the ignitability of the air-fuel mixture deteriorates, resulting in a decrease in engine output, etc., so EGR
The rate must be increased or decreased according to engine operating conditions.

The EGR rate is increased or decreased by opening or closing a recirculation amount control valve (hereinafter referred to as an EGR valve) provided in an EGR gas passage that communicates the exhaust system and the intake system. Conventionally, as a method for controlling the opening and closing of the EGR valve, pneumatic control using intake pipe negative pressure and exhaust pressure has been generally performed. However, due to the fact that pneumatic control cannot perform complex control over various engine operating conditions, the demand for exhaust gas purification has increased over the past year, and the development of centralized engine control systems, electronic control systems are now being used. has become mainstream.

An electronically controlled EGR device adjusts the EGR rate to a pre-programmed EGR rate depending on the engine speed, load, etc.
The GR valve is opened and closed.

In addition to negative pressure, a pulse motor or the like is also used as a driving source for the EGR valve, and feedback control is also performed in many cases. Below, as an example of an electronically controlled EGR device,
The configuration and operation of the EGR valve lift feedback system and the intake pipe 0° sensor output feedback system will be briefly described.

What is the EGR valve lift feedback system?
As shown in Publication No. 0-11214, this method includes a drive means for driving the EGR valve to open and close, and a detection means for detecting the opening amount of the EGR valve, and the EGR valve is operated according to the operating state of the engine. The valve is driven and feedback control is performed to maintain a predetermined valve opening amount.

As the driving means, an intake pipe negative pressure, a pulse motor, etc. are used, and as the valve opening amount detecting means, a bulge meter, etc. is used.

Also, what is the intake pipe 02 sensor output feedback method?
As shown in Japanese Patent Application Laid-Open No. 60-138264, this system includes a drive means for opening and closing the EGR valve and an 02 sensor provided in the intake pipe. In this method, the EGR valve is driven according to the operating state of the engine, the actual EGR rate is calculated using the oxygen concentration measured by the 0□ sensor, and the deviation from the target EGR rate is detected to perform feedback control. be. As the 02 sensor, a solid electrolyte oxygen pump type oxygen concentration measuring device or the like is used.

<Problems to be Solved by the Invention> However, each of the above-mentioned systems had the following drawbacks.

In the EGR valve lift feedback method, durability is poor because a large acceleration is applied to the valve opening detection means intermittently during engine operation. Since the system detects the opening amount of the EGR valve instead of detecting the opening amount of the EGR valve, even if the actual EGR rate changes due to wear of the EGR valve, carbon deposition, etc., correction cannot be made for this change.

In the 02 sensor output feedback method, the actual EGR
Since feedback control is performed using the EGR rate, control accuracy under constant load conditions is excellent, but there is a time lag between when the EGR valve operates and when the actual EGR rate is measured, so hunting may occur and transient There was a problem with the time response. Therefore, while the driving force of the EGR valve, such as negative pressure and motor output, is calculated based on the driving force to start opening the EGR valve, control is performed using a proportional integral method or the like.

However, this driving force to start opening the valve itself is not constant due to differences between individual EGR valves and changes over time such as carbon deposition, which hinders accurate control. Additionally, when EGR gas is not recirculated, such as during idling, it is common practice to drive the EGR valve with the opening driving force to improve responsiveness. If the value is not accurate, responsiveness may deteriorate or unnecessary EGR may occur.

Means for Solving the Problems> Therefore, in the present invention, a part of the exhaust gas of an internal combustion engine is introduced into the intake system to lower the combustion temperature, thereby reducing the concentration of nitrogen oxides in the exhaust gas. The recirculation device includes: an operating state measuring means for measuring the operating state of the internal combustion engine; a recirculation amount control valve that is provided in the exhaust gas recirculation passage and increases or decreases the amount of recirculation by opening and closing the passage; A control valve driving means for driving a circulation amount control valve, a gas concentration detector for detecting an amount of exhaust gas recirculation, and a target amount of exhaust gas recirculation is determined based on the measurement results of the operating state measuring means. Then, the control valve driving means is used to control the opening and closing of the recirculation amount control valve so that the opening amount corresponds to the target exhaust gas recirculation amount, while the exhaust gas is controlled based on the operating conditions of the internal combustion engine. When a state in which no recirculation continues for a certain period of time, the control valve driving means is used to gradually apply a driving force to the recirculation amount control valve, and the gas concentration detector determines when to start exhaust gas recirculation. The recirculation amount control valve is provided with a control means for learning and correcting the data of the driving force for opening the recirculation amount control valve based on the detected values.

Function> In the exhaust gas recirculation device according to the present invention, after determining the recirculation amount based on the measurement result of the operating state measuring means, the recirculation amount control valve is driven using the control valve driving means to control the exhaust gas. While performing gas recirculation, the opening amount of the recirculation amount control valve is gradually changed during periods such as idling when exhaust gas recirculation is stopped, and a gas concentration detector detects the timing to start recirculation.
Learning and correcting the data of the driving force to start opening the valve.

Embodiment An embodiment in which the present invention is applied to an engine equipped with an electronically controlled fuel injection device (hereinafter referred to as ECI) will be specifically described with reference to the drawings.

Fig. 1 schematically shows the entire central control system in this embodiment, and Fig. 2 shows the eight-domain configuration of the electronic control unit (hereinafter referred to as ECU) which is the control center of the central control system. . Further, FIGS. 3 and 4 show control flowcharts of the main routine and idle learning routine, respectively, in this embodiment. FIG. 5 shows the target EGR rate, FIG. 6 shows the flow rate characteristics of the EGR valve with respect to the control negative pressure, and FIG. 7 shows the duty ratio of the solenoid valve.

As shown in FIG. 1, the engine E includes an injector 1 for injecting fuel, a spark plug 2 for ignition, and an EG
The opening degree of the R valve 3 is entirely under the control of the ECU 4. The ECU 4 drives and controls these controlled devices (injector 12, spark plug 2, EGR valve 3) based on various data. Below, an overview of the sensors used by the ECU 4 to collect various data and the operation of controlled devices will be described along the flow of intake air.

The air sucked under negative pressure from the air cleaner 5 by the descent of the piston E1 in the engine E is transferred to a Karman vortex type air flow meter 6, an intake air temperature sensor 7 and an atmospheric pressure sensor 8.
The intake air amount, intake temperature, and atmospheric pressure are detected.

The amount of air flowing into the intake pipe 9 is controlled by a butterfly-type throttle valve 10.
The opening degree is detected by the throttle sensor 11.

A negative pressure take-off pipe 12 and an EGR gas introduction pipe 13 are connected and opened in this order to the intake pipe wall downstream of the throttle valve 10, and air flows from the negative pressure take-off pipe 12 and EGR gas flows from the EGR gas introduction pipe 13. Each of them is designed to be sucked into the intake pipe 9. The intake air is mixed with this EGR gas, and the oxygen concentration is detected by the 02 sensor 14 provided downstream. The 0□ sensor 14 is similar to that described in the prior art section.

Injectors 1 for the number of cylinders are provided near the end of the intake pipe 9 on the combustion chamber side, and are opened in response to a command from the ECU 4 to inject the amount of fuel required by each cylinder.

Fuel is pumped from a fuel tank (not shown) to the injector 1 by a fuel pump (also not shown).

In the figure, 15 is a water temperature sensor, which detects the cooling water temperature.

Fuel is injected from the injector 1, and the air that has become a mixture is drawn into the combustion chamber E2, and is ignited by the ignition plug 2 near compression top dead center. In Figure 1, due to space limitations,
Although the spark plug 2 is shown in a distant position, the tip of the spark plug 2 naturally faces into the combustion chamber E2.

High voltage from an ignition coil 16 is sent to the ignition plug 2 via a distributor 17. The power transmission timing (ignition timing) is calculated within the ECUA, and based on the calculation result, the power is transmitted to the ignition coil 1 via the power transistor 18.
6 is being driven. A crank angle sensor 19 and a cylinder discrimination sensor 20 are built into the distributor 17 to detect the rotational state of the engine.

When the explosion and expansion strokes are completed, the mixture becomes exhaust gas, most of which passes through the exhaust pipe 21 and passes through the catalytic converter 21.
After the harmful components are burned in a, they are released into the atmosphere from a muffler (not shown).

A part of the exhaust gas is then introduced as EGR gas to the EGR gas extraction IW 22, the amount of which is controlled by the EGR valve 3, and introduced into the intake pipe 9 from the above-mentioned EGR gas introduction v:13.

The EGR valve 3 is operated by the negative pressure from the negative pressure outlet pipe 12 described above, and this negative pressure is increased or decreased by opening and closing an ON-OFF type solenoid valve 23 provided in the line of the negative pressure outlet pipe 12. It looks like this. In Fig. 1, 24 indicates a filter 2 at the end.
5 is attached and an orifice 24a is provided in the pipe line. This communication pipe 24 is for gradually introducing atmospheric air into the EGR valve 3 to close the EGR valve 3 when the passage of the negative pressure extraction pipe 12 is blocked by the solenoid valve 23. The solenoid valve 23 is opened and closed by commands from the ECU 4, and its control method is a variable duty ratio type.

The ECU 4 controls the EG according to various operating conditions of the engine E.
Change the duty ratio of the solenoid valve 23 so that the R rate is achieved.
Controls the opening amount of the GR valve 3. In addition, in Figure 1, 26.2
Reference numerals 7 and 28 denote a battery as a driving power source for the ECU 4, a battery sensor for detecting battery voltage, and an ignition switch, which are also connected to the ECU 4.

As mentioned above, various sensors and controlled devices are connected to the ECU 4, but the 8-domain of the ECUA itself is the second one.
As shown in the figure, it is mainly configured with a CPU (central processing unit) 29. Since the data from the intake temperature sensor 7, atmospheric pressure sensor 8, throttle sensor 11, 02 sensor 14 and battery sensor 27 are analog signals,
C via interface 30 and A/D converter 31
It is input to PU29. Data from the ignition switch 28 is input to the CPU 29 via an interface 32, and data from the crank angle sensor 19, cylinder discrimination sensor 20, and air flow meter 6 are input directly to the CPU 29.

The CPO 29 also has a ROM (read only memory) 33. through the pass line. RAM (rewrite memory) 3
4 and BUR, which stores the memory contents even if the ignition switch 28 is turned off while the battery 26 is connected.
Data is exchanged with AM (backup memory) 35.

Inside the CPO 29, the fuel injection amount, the ignition timing, and the opening degree of the EGR valve are determined using the above-described various data and memory. Then, the injector 1 is driven through the injector driver 36, the power transistor 18 is driven through the ignition driver 37, and the solenoid valve 23 is driven through the EGR driver 38.

The operation of this embodiment will be described below with reference to FIGS. 3 to 7. Prior to a detailed explanation, an outline of the division in this embodiment will be briefly described.

In this embodiment, first, the target EGR rate is determined from the measurement results of the driving state measuring means, and the actual EGR rate is determined from the detection results of the 02 sensor. Then, from these deviations, the drive duty ratio of the EGR valve is calculated and determined by the proportional integral method, and the EGR valve is
Drive the valve. This calculation is performed based on the valve opening start negative pressure (hereinafter referred to as valve opening negative pressure) specific to the EGR valve.

On the other hand, when idling continues for 30 seconds, learning starts. In this case, the O2 sensor detects the oxygen concentration in the intake air while increasing the drive duty ratio step by step after setting it to 0.

When recirculation of EGR gas is detected, the opening negative pressure of the EGR valve is corrected based on the drive duty ratio at that time.

In this way, unlike the conventional 02 sensor output feedback method described above, this embodiment has a function of learning and correcting the EGR valve opening start driving force (N valve negative pressure), so responsiveness is improved. At the same time, the effects of changes over time are also reduced.

Hereinafter, the flow of control will be specifically described in accordance with the flowcharts of FIGS. 3 and 4.

When engine E starts, EGR @ control is also started at the same time. The control shown in the flowchart of FIG. 3 is activated by a signal (crank interrupt signal) from the crank angle sensor 19, and is executed every crank rotation.

When the control is started, the CPU 29 detects the crank angle sensor 19.
From the output signal of the air flow meter 6, the engine rotational speed Ne and intake air ff1A are detected, and then the load A/N is calculated based on these detection results. next,
Call up the EGR rate map (map Ml shown in Figure 5) stored in the ROM33, and enter the rotation speed Na and load A/
Determine the target EGR rate (EGR) T from N.

Next, the output voltage IP of the 0□ sensor 14 is detected, and this output voltage ■2 is corrected by the load A/N. This is because the output characteristics of the 02 sensor 14 change depending on the intake pipe negative pressure (proportional to the load A/N). Next, the actual EGR rate (EGR) A is determined from the output current value {circle around (2)}. For calculations, F, G
The output current value of the 02 sensor 14 when R is not performed (I
P).

is used. (EGR)A is given by the following formula.

Next, the target EGR rate (EGR) T and the actual EGR rate (EGR
Find the deviation Ee from the input using the formula below.

Ee= (EGR), (EGR)A-(21st order,
Find the proportional gain for proportional control using the following procedure.

First, calculate the instantaneous proportional gain (KP) from the engine rotation 1iiNe and the load A/N using the formula below. Find 8.

(K,,),,=(KP),NXf(A/N)XNe
・(3) Here, (Kp)IN is the proportional gain base data (fixed value) stored in the ROM33, and f
(A/N) is a correction function.

Next, this instantaneous proportional gain (KP). After comparing the instantaneous proportional gain (Kp)'oFI of the previous time (before one revolution of the crank) with the instantaneous proportional gain (Kp)'oFI, the proportional gain salt is determined using an arithmetic expression according to the result.

(Kp). , ≧(Kp)'oR, then Kp=KPUXK,,'+ (1-KPU) x(K
p). , −(41(Kp)cp+<(Kp)
If 'on', Kp=9DXKp'+(1-KPD)X
(Kp)C+t...(5] Here, KPU, Kp
D is the smoothing coefficient (
0≦KPU<1.

0≦KD<1), and is a fixed value stored in the ROM 33. Further, KP' is the previous proportional gain.

Next, the integral gain K for performing integral control.

seek.

First, the instantaneous integral gain (KI) CFI is obtained by the following formula using the load A/N.

(K,). , = (K,), Nxf (A/N)
-(61) Here, (Kl)IN is integral gain base data (fixed value).

Next, this instantaneous integral gain (KI) C11 is compared with the previous instantaneous integral gain (Kl)'CR, and the proportional gain KP is
Obtain the integral gain using the same procedure as .

If (KI)cR≧(Kl)′CR, then U
XK, '+ (1-to, U) X(K, ). 8 ・
...(7) (Kl) CI'l < (Kl '' CR, then =, DXK, '+ (1-, D) X
(K, ). , -f8) Therefore, K, LJ, K,
D is a smoothing coefficient (0≦, U<1, 0≦, D<1) when the integral gain increases and decreases, which are fixed values, respectively, and K is the previous integral gain.

Next, the integral gain is given as the deviation Ee obtained by equation (2).
Therefore, the addition data (EG
R).

(EGR), = to, XEe/Kl −
(9) K1 is a sufficiently large constant to stabilize control, that is, to prevent hunting.

Next, the integral data (EGR) is updated using this addition data (EGR).

(EGR),, =(EGR)',, +(EGR),
...(11 Here, (EGR)', is the integral data up to the previous time.

](++In the calculation of the n formula, the upper and lower limit values of (EGR), , are determined, and the upper and lower limit values are set so as not to go outside the range.

Next, proportional gain KP, (l difference Ee and integral data (
feedback data (EGR), p, which is the target EGR flow rate, is obtained from EGR), , .

(EGR), p=KPxEe+(EGR),
++・(ii) Next, this feedback data (E
GR) + pb' et al., calculate the converted flow rate (E G '') IITO at -200 Kg Hg.

This is usually - when measuring EGRGaO2 amount characteristics.
This is because the test is performed by applying a negative pressure of 200 nwaHg. (EGR) STo can be obtained from the exhaust pressure P,...h (as it is almost the same as atmospheric pressure, the input data of the atmospheric pressure sensor 8 is used in the example) and the intake pressure P (calculated from the load A/N). It is given by the following formula using the intake pipe negative pressure P-P8.

Although the intake pressure P8 was determined from the load A/N in this embodiment, if an intake pressure sensor is provided, the value of the intake pressure sensor is of course used.

Next, the EGRGaO2 control negative pressure P required to ensure the converted flow rate (EGR) sTo is determined. As shown in map M2 in FIG. 6, the EGRGaO2 amount characteristic used in this example is linear with respect to the control negative pressure, and the valve opening start negative pressure (hereinafter referred to as valve opening negative pressure)
The full-open negative pressure is PEl and P52, respectively, and the full-open flow rate is (EGR).

Control negative pressure P5 is given by the following formula.

Here, K2 is a correction coefficient.

Next, a comparison is made between Pl: and P,2. This is (11
As a result of the calculation using the formula, the control negative pressure Pl: may become larger than the fully open negative pressure P,2, and if this is used as it is, the response when closing the valve will deteriorate. (EGR)sTo≧(EGR) -1, then P! , =
PE2... (Set as 141.

Then, the duty ratio corresponding to the control negative pressure PI:□ and the negative pressure A/N is searched from the map (map M3) in FIG. 7, and the solenoid valve 23 is driven with this duty ratio.

If p < p, then EGRGaO2 chain mode <
(EGR) Determine whether svo = Q'. The range in which EGR is not performed is an operating range such as idle, full load, and high rotation, and is determined from signals from an idle switch (not shown), air flow meter 6, etc. and,
If it is not the closed mode, the duty ratio is searched from the map using P6 obtained using equation 11, and the solenoid valve 23 is driven.

On the other hand, if the vehicle is in the closed mode, it is then determined whether the vehicle is in an idling state. If the vehicle is not in an idling state, the duty ratio is set to 0%, that is, the driving of the solenoid valve 23 is stopped. When the driving of the solenoid valve 23 is stopped, as shown in the map of FIG. 7, the control negative pressure P6 becomes the valve opening negative pressure P6. becomes smaller, EGRGaO2
The body is pressed against the valve seat.

The reason for doing this is that the pulsating negative pressure in the intake pipe increases in operating ranges other than idling, that is, full load and high rotation, so the control negative pressure P6 opens the valve negative pressure Pl! This is because if it is set to 1, the EGR GaO2 body will cause surging and unnecessary EGR gas will be introduced.

In the case of idle operation, an idle learning routine activation signal (hereinafter referred to as an idle activation signal) is transmitted, and the fourth
The idle learning routine shown in the flowchart of the figure is activated.

This routine is started by the main routine shown in FIG. 3, but after being started, it is operated by a timer independently of the main routine.

When the operating state is other than idle, the process is stopped at that point. Note that while the idling operation continues, the idling start signal is transmitted every time the crank rotates, but after the idling is started, the idle learning routine performs processing regardless of the start signal.

The processing in the idle learning routine will be explained below.

When this routine is started, first set the control negative pressure to PE"PE1...4, search the duty ratio from the map, and check the solenoid valve 23.
Drive for 30 seconds. This is because when the pulsating negative pressure in the intake pipe is small, no surging occurs in the EGRGaO2 body even when the control negative pressure P6 is used as the valve opening negative pressure P6□, and the responsiveness when opening the valve is improved. Further, the reason for driving in this state for 30 seconds is to stabilize the EGRJJ position.

Next, after 30 seconds have passed in that state, the duty ratio is set to D=O%...(II) for 3 seconds.Then, after that, 02 sensor 1
Measure the output current value lp of 4, and calculate the actual EG using equation (1).
Calculate the R rate (EGR) A. Here, the reason why the timing is 3 seconds is the time required for EGRGaO2 gas to be introduced.
This takes into consideration the distance between the EGR GaO2 sensors 14.

Next, determine whether the actual EGR rate (EGR) A is greater than 1%, and if (EGR) A is 21%, obtain the valve opening negative pressure data P0 from the map M3. At this time, the decision criterion was set to 1% in consideration of the accuracy of the 02 sensor 14 and the adverse effects of the introduction of EGR gas.

Once the valve opening negative pressure data 8 is obtained, the valve opening negative pressure P can be calculated using the formula below.
Calculate and update El.

PE, = -P, + (1-k) P', □ ...
(1?l Here, ``it'' is the previous valve opening negative pressure, and k is a filter coefficient set sufficiently small to prevent hunting.

On the other hand, if (EGR)A<1%, the solenoid valve is driven for 3 seconds with the duty ratio set to D=2%...α. In this case as well, it is determined whether the actual EGR rate (EGR) A is greater than 1%, and if (EGR) A is 21%, the valve opening negative pressure data P is obtained. (Calculate and update the valve opening negative pressure P6 using equation 1.

If (EGR)A<196, the duty ratio is further set to 4%, the solenoid valve is driven, and a determination is made.

If (EGR)A<1%, control and measurement are performed by increasing the breath duty ratio by 2% until (EGR)A reaches 21%.

The valve opening negative pressure Pl, :l corrected in this way is used in formula (131) in the main routine described above, and is used in the EG
This improves the accuracy of R control.

Further, even while this idle learning routine is operating, calculations are performed every one rotation of the main routine (from rank to rank), and EGR control is performed by the main routine from the moment the engine enters a non-idling operating state.

This concludes the explanation of this embodiment, but it goes without saying that the present invention is not limited to this embodiment.
The present invention may be applied to an engine having a carburetor instead of I. Further, the recirculation amount detection means may be a device other than the 02 sensor, or an EGR valve having a high-order curved flow rate characteristic or driven by a pulse motor or the like instead of negative pressure may be used. . Regarding control, in this embodiment, the present invention was applied to the 02 sensor output feedback method in which the control negative pressure is calculated based on the valve opening negative pressure. May be applied.

In addition, although the calculation method and calculation procedure in the control were specifically described in the embodiment, the EGR valve was sequentially driven and
It is sufficient if the valve opening driving force is detected by a sensor or the like and the learning correction is performed, and there is no need to set the duty ratio to 0% and then add it as in the example, and the 30 second waiting period can be applied. There is no need to set a time.

<Effects of the Invention> In the EGR device according to the present invention, the valve opening driving force, which is the standard for EGR control, is corrected by learning, so the optimal EGR control is performed by the amount. As a result, environmental pollution due to harmful exhaust gases is reduced.

[Brief explanation of the drawing]

FIGS. 1 and 2 are a schematic diagram of a centralized control system and a hardware configuration diagram of an electronic control unit, respectively, in an embodiment of the present invention. FIGS. 3 and 4 are control flowcharts explaining the operation of the embodiment. Figures 5 to 7 are
These are maps of the target EGR rate, the flow rate characteristics of the EGR valve, and the duty ratio of the solenoid valve. In the figure, 3 is the EGR valve and 4 is the ECU. 14 is an O sensor, 23 is a solenoid valve, 33 is ROM1, and 34 is RAM. 35 is BURAM. Figure 5 Target EGR rate M, A (Ma1.'M +) Engine speed Ne Figure 6

Claims (1)

[Scope of Claim] An exhaust gas recirculation device that reduces the concentration of nitrogen oxides in the exhaust gas by introducing a portion of the exhaust gas of the internal combustion engine into the intake system to lower the combustion temperature, comprising: a recirculation amount control valve that is provided in the exhaust gas recirculation passage and increases or decreases the amount of recirculation by opening and closing the passage; and a control that drives the recirculation amount control valve. a valve driving means, a gas concentration detector for detecting the amount of exhaust gas recirculation, and a target amount of exhaust gas recirculation based on the measurement results of the operating state measuring means; The control valve driving means is used to control the opening and closing of the recirculation amount control valve so that the valve opening amount corresponds to the recirculation amount, while a state in which exhaust gas recirculation is not performed is constant as determined by the operating conditions of the internal combustion engine. When the time continues, the control valve drive means is used to apply a gradual driving force to the recirculation amount control valve, and the gas concentration detector detects the start timing of exhaust gas recirculation. An exhaust gas recirculation device characterized by comprising: a control means for learning and correcting data of a driving force for opening a circulation amount control valve.