JPH01290922A - Method for controlling supercharging pressure of internal combustion engine - Google Patents

Abstract

PURPOSE:To improve the accuracy of a controlled variable by renewing a learned value to a value obtained at the time of feedback control, at the time of the open-loop control of supercharging pressure. CONSTITUTION:Supercharging pressure is controlled by driving the actuator 17 of the movable vane of a supercharger 4 by an electromagnetic control valve 18. A feedback control is carried out when the supercharging pressure is in a steady condition. At the time of a transient condition, an open-loop control is carried out based on a learned value which is calculated at the time of the feedback control. This learned value is renewed to a value obtained by learning for each of a plurality of zones which are determined by intake air temperature and engine speed at the time of the feedback control. Thereby, it can be prevented that a value which is learned at the time of a different target pressure is applied at the time of the open-loop control, enabling a controlled variable to be accurately set.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to a method for controlling the boost pressure of an internal combustion engine, and particularly to a method for controlling the boost pressure of an internal combustion engine to appropriately control the boost pressure during a transient state. Concerning a control method.

(Prior Art and Problems to be Solved by the Invention) In a supercharged internal combustion engine capable of controlling boost pressure in a so-called variable displacement turbocharger that uses an exhaust gas flow as a turbine drive source, an appropriate actuator is used. Feedback control is used when the boost pressure is controlled by a pressure-responsive actuation system that includes a diaphragm that uses boost pressure or negative pressure in the intake pipe as the operating pressure, or an actuation system that uses a step motor or the like.

This is because supercharging pressure is influenced by individual differences such as variations in manufacturing of the actuator and the components of the turbocharger body, and changes over time such as deterioration of durability due to use. When controlling boost pressure to a desired value, operating with a preset control amount will result in a large change in boost pressure (even if the same control limit is applied, the In some cases, the amount of adjustment of the actuation control system such as the actuator may differ due to differences between individuals as mentioned above or due to changes over time, so feedback control of boost pressure is used. For example, this is because this can be easily resolved. In other words, feedback control detects the deviation between the target boost pressure and the actual boost pressure and determines the control amount according to the deviation so that the deviation becomes zero. Even if there are variations in the measurements of the actuators, or even if there are changes over time, these effects will be absorbed and corrected in this control.

Therefore, when controlling the boost pressure to the required target pressure corresponding to the engine operating condition, the control amount is determined based on the above deviation so that the boost pressure reaches the target pressure, and the boost pressure is maintained at the target pressure. Although it is desirable to perform feedback control to control the amount of fuel, feedback control may cause an undesired abnormality or rise in boost pressure depending on the case.

In other words, in a so-called filtration transient state, such as when the boost pressure suddenly increases, the control fJg amount cannot follow the boost pressure adjustment due to the response delay of the control system, resulting in an abnormal increase in the boost pressure or an abnormality. Phenomena such as lowering and hunting occur. Therefore, in order to eliminate the above-mentioned problems with feedback control in transient states, the applicant set the boost pressure control to oven loop control in transient states, and changed it to steady state. Next, we will propose a method for controlling boost pressure using feedback control.

(Japanese Patent Application No. 61-275783), and according to this, it is possible to perform more stable supercharging pressure control than the previous one.

On the other hand, in such a supercharging pressure control method, paying particular attention to the oven loop control, in order to compensate for the shift in the supercharging pressure control during the oven loop control, the control amount of the supercharging pressure during the open loop control is In addition to setting based on the learned value calculated during feedback control, the learned value is calculated separately for each region according to the intake temperature, paying attention to the fact that the intake temperature does not reach the boost pressure and the impact is large. In other words, the applicant has already proposed a method in which the learning value is updated for each of a plurality of different intake air temperatures to perform more appropriate boost pressure control (Japanese Patent Application No. 61-275784). No. etc.).

However, in the case of this method of controlling claw correction during open-loop control, the intake temperature correction learned value is divided and stored for each intake temperature range during feedback control, and then read and applied during open-loop control. In order to appropriately control the supercharging pressure during loop control, that is, in a transient state, there is still room for improvement in the following points.

In other words, the feedback control region in which the learning value is calculated is the steady state of the boost pressure, but even if the region corresponding to the intake air temperature is the same in the steady state, the state of the boost pressure is not necessarily the same. In particular, the optimal boost pressure is always set regardless of the intake temperature and engine speed. If control is performed to set a target value of the boost pressure, the state of the boost pressure will also change as the target value is controlled to change. Therefore, if you simply divide and rewrite (update) the learned value for intake temperature correction for each intake temperature, the value obtained by learning at a different target pressure may be mistakenly rewritten. There will be cases where it is stored away. If it is rewritten incorrectly, the learned value itself applied during open-loop control will deviate, making it impossible to set the control amount of boost pressure using the appropriate learned value during open-loop control, and this will result in the overload at that time. A deviation occurs in the boost pressure, and therefore, the boost pressure cannot be controlled appropriately.

The present invention is intended to further improve the above-mentioned problems, and even when changing the target value of boost pressure to an optimal value depending on the intake temperature and engine speed,
Boost pressure of an internal combustion engine that can always obtain an appropriate learned value and can appropriately control boost pressure even during open loop control in which the learned value is applied to execute boost pressure control. The purpose is to provide a control method for

(Means for Solving the Problems) In order to achieve the above object, the present invention provides a method for controlling supercharging pressure of an internal combustion engine, which controls supercharging pressure by controlling a control claw of a supercharging pressure control means. The target boost pressure is set according to the intake air temperature and the engine speed, and the control amount is fed back according to the deviation between the actual boost pressure and the target boost pressure when the boost pressure is in a steady state. control, and when the boost pressure is in a transient state, open loop control is performed based on the learned value calculated during the feed pump nose control, and the learned value is used to control the intake air temperature and engine rotation speed during the feedback control. The value is updated to the value obtained by learning for each of the multiple areas determined by .

(Example) Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is an overall configuration diagram of a control device for an internal combustion engine equipped with a turbocharger to which the control method of the present invention is applied.

Reference numeral 1 in the figure indicates, for example, a 6-cylinder internal combustion engine,
An intake pipe 2 is on the upstream side of the engine 1, and an exhaust pipe 3 is on the downstream side
A turbocharger 4 is interposed between the intake pipe 2 and the exhaust pipe 3.

The intake pipe 2 is provided with an air cleaner 5, an intercooler 6, and a throttle valve 7 in this order from the second-stream side.

An atmospheric pressure (P^) sensor 8 is installed between the air cleaner 5 and the turbocharger 4 at the open end of the intake pipe 2 on the atmosphere side, and detects the intake pressure in the intake pipe section between them, that is, the atmospheric pressure. , and its detection signal is supplied to an electronic control unit (hereinafter referred to as rEcUJ) 9.

A throttle valve opening (Oro) sensor 10 is connected to the throttle valve 7 to convert the valve opening of the throttle valve 7 into an electrical signal and send it to EC[J9.

On the other hand, the intake pipe absolute pressure (P
BATC) sensor 11 is provided, and this P BA
The absolute pressure signal converted into an electrical signal by the TC sensor 11 is sent to the ECU 9.

In this embodiment, the PBATC sensor 11 executes feedback control in the substantially fully open region of the throttle valve 7 during boost pressure control by the turbocharger 4, as will be described later. The magnitude of the boost pressure can be detected as the intake pressure value in the intake passage obtained by the PBATC sensor 11 provided downstream of the throttle valve 7.
9 is also supplied with information regarding the boost pressure from the P BATC sensor 11 under the condition that the throttle valve is fully open.

In addition, downstream of the P BATC sensor sensor, the intake air temperature (T^
) A sensor 12 is attached, which detects the intake air temperature T^ and outputs a corresponding electric signal to supply it to the ECU 9.

A fuel injection valve 13 is provided in the intake pipe 2 between the engine 1 and the throttle valve 7. This fuel injection valve 13 is provided for each cylinder slightly upstream of the intake valve of the intake pipe 2 (2
each injection valve 13 is connected to a fuel pump (not shown) and electrically connected to the ECU 9.
A signal from the ECU 9 controls the opening time of the fuel injection, that is, the amount of fuel supplied.

The engine rotation speed (N e ) sensor 14 is connected to the engine 1
It is attached around the camshaft or crankshaft (not shown) and connected to the ECU 9, and outputs a TDC signal, that is, one pulse at a predetermined crank angle position every 180° rotation of the crankshaft of the engine l. ECU
Supply to 9.

Further, a three-way catalyst 15 is arranged downstream of the turbocharger 4 in the exhaust pipe 3 .

The turbocharger 4 is of a variable capacity type, and its operation control system includes an actuator 17 having a drive rod 16 linked to a movable vane 6 (see FIG. 3), which will be described later, and a duty-controlled supercharging system. It has a pressure introduction electromagnetic control valve 18 (hereinafter simply referred to as "control valve J") and a regulator 19.

FIG. 2 shows an overall configuration diagram of the turbocharger 4. That is, the turbocharger 4 has a compressor casing 41 forming a scroll of the compressor part, a back plate 42 closing the back surface of the compressor casing 41, etc., which pivotally supports the main shaft 43 of the turbocharger 4, and lubricates the bearing. It has a bearing casing 44 that incorporates a structure in which cooling water circulates, and a turbine casing 45 that forms a scroll of the turbine portion.

Inside the compressor casing 41, there are formed a scroll passage 46 and an axial passage 47 to which the intake pipe 2 is connected, the former 46 serving as an intake outlet, and the latter 4
7 represents the intake population.

In the intake l21 section, a compressor wheel 48 is attached to one end of the main lqb43.

The bearing hole 49.50 of the bearing casing 44 is provided with a radial bearing metal 51 and a thrust l-bearing metal 52.
The main shaft 43 is pivotally supported by.

Further, the bearing casing 44 is formed with a lubricating oil introduction hole 53, a lubricating oil passage 54, a lubricating oil outlet 55, and a water jacket 56.

Inside the turbine casing 45, a scroll passage 5 is provided.
7, and the person's 11 opening 57a opening their eyes in the tangential direction.
An outlet passage 58 extending in the axial direction and an outlet opening 58a thereof are formed, and the inlet opening 57a and the outlet passage 58a are each connected to the exhaust pipe 3.

At the outer periphery of the fixed vane member 6° fixed to the back plate 59 so as to be disposed in the center of the scroll passage 57,
As shown in FIG. 3, a plurality of fixed vanes 62, for example, four fixed vanes, are formed so as to concentrically surround a vane wheel 61 provided at the other end of the main shaft 4:3. ing. These fixed vanes 62 each have a partial arc shape, and are provided with the same width and at equal intervals along the circumferential direction. A movable vane 64 is arranged between each of the fixed vanes 62, which is rotatably pivoted to the back plate 59 and fixed to the free end of a rotating bin 63.

These movable vanes 64 have an arc shape with the same curvature as the fixed vane 62, and are located at approximately the same circumference, and have a minimum opening position shown by a solid line in FIG. It can be rotated between the fully open position shown.

The gaps between the fixed vanes 62 are opened and closed by the movable vanes 64 being rotated in synchronization, and the circulation area of each gap corresponds to the inclination angle of the rotating claws, that is, the movable vanes 64. will be adjusted accordingly.

The synchronized rotational drive of each movable vane 64 is achieved by a rotary bin 63 that supports each vane, a link mechanism 65 (FIG. 2) connected to the rotary bin 63, and a link mechanism 65 that is connected. This is done by the actuator 17 via the previously described drive rod 16 (FIG. 1). The drive rod 16 and the link mechanism 65 are configured such that when the drive rod 16 is operated in the extension direction (to the left in FIG. 1), the opening degree of each movable vane 64 increases and the circulation area of each gap increases. When operated in the contraction direction (rightward in Figure 1), the opening is reduced and the flow area of each gap is reduced, and by controlling the opening, the turbo The capacity of charger 4 is adjusted.

That is, in the II) variable displacement turbocharger 4 having the above configuration, the exhaust gas discharged from the engine 1 body is
The movable vane 6 flows into the scroll passage 57 from the turbine side inlet passage 57a, and the movable vane 6 flows in accordance with the amount of rotation of the movable vane 64.
Exhaust gas flows into the turbine wheel 61 at a flow rate according to the flow area of the gap between the turbine wheel 4 and the fixed vane 62, rotates the turbine wheel 61, and is discharged from the outlet passage 58. The flow velocity of the exhaust gas that drives the turbine wheel 61 is
As a result of dependence on the above-mentioned gap circulation area, if the circulation area of the gap between each movable vane 64 and the fixed vane 62 is small and the flow velocity is high, the rotation speed of the turbine wheel 61, that is, the main shaft 43 becomes high, and each movable vane 64 and the fixed vane If the flow area of the gap between the vanes 61 is large and the flow velocity is low, the turbine wheel 6
1, that is, the rotational speed of the main shaft 43 becomes slower. Since the compressor wheel 48 rotates in accordance with the rotation of the turbine wheel 61, the air led from the air cleaner 5 to the axial passage 47 is compressed by the compressor wheel 48 according to its rotational speed, and passes through the scroll passage 46 to the intercooler. 6, and the intake air is pressurized.

Thus, when the movable vane 64 is positioned at the outermost (J) position in the radial direction of the turbine casing 57 and the air gap circulation area between it and the fixed vane 62 is minimized, that is, the opening degree is minimized, the supercharging pressure is maximized, and the movable vane 64 is When the vane 64 is located at the outermost position in the radial direction of the turbine casing 57 and the air gap circulation area between it and the fixed vane 62 is maximized, that is, when the opening is maximized, the supercharging pressure is minimized, and the movable vane 64 is opened. A high boost pressure can be easily obtained by adjusting the opening degree, and the boost pressure can be changed by adjusting the opening degree in a wide range between the minimum and maximum opening degrees. .

1. The actuator I7, which rotationally drives the movable vane 64 of the charger 4 for supercharging pressure control, operates between a first pressure chamber 17b defined by a diaphragm 17a and a second pressure chamber, as shown in FIG. The drive rod 16 described above passes through the housing on the second pressure chamber 1'7c side and is connected to the diaphragm 17a. Second
A spring 17d inserted into the pressure chamber 17b biases the diaphragm 17a in the direction in which the drive rod 16 contracts, that is, in the direction in which the movable vane 64 reduces the opening degree.

An intake passage between the air cleaner 5 and the turbocharger 4 is connected to the first pressure chamber 17a via a throttle 22, and an intake passage between the turbocharger 4 and the intercooler 6 is connected to the regulator 19, the throttle 23, and the control valve 18. connected via.

The control valve 18 is a normally closed on-off two-position operating solenoid valve, and includes a solenoid 18a and a valve body 18b that opens when the solenoid 18a is energized. When the valve body 18b is opened by energizing the solenoid 18a, the supercharging pressure in the intake passage between the turbocharger 4 and the intercooler 6 is applied to the first pressure chamber 17b of the actuator 17.
will be introduced in

In this case, the diaphragm 17r is deflected so as to extend the drive rod 16, and the movable vane 64 of the turbocharger 4 is moved inward via the drive rod 16 and the link mechanism 65, that is, its opening is increased. When the valve body 18b is closed, the introduction of supercharging pressure is cut off, and the movable vane 64 is driven in the direction in which its opening degree becomes smaller, contrary to the above description. .

Therefore, the valve body 18b in one cycle of the on-off of the solenoid 18a, that is, the opening and opening of the valve body 18t).
The ratio of the closing time of , that is, the valve closing duty ratio Dva (hereinafter,
111 (referred to as the "duty ratio"), this changes from the 100% state (the state where the movable vane 64 is at the minimum opening position described above and the maximum boost pressure is reached) to supercharging according to the duty ratio Dvc. The magnitude of the pressure is controlled.

The solenoid 18a of the control valve 18 is connected to the ECU.
9, and the duty ratio Dva is controlled by a signal from the ECU 9.

Furthermore, E C1 and J 9 have vehicle speed (V) for detecting vehicle speed.
A sensor 24 is connected and its detection signal is supplied.

EC1J9 shapes the input waveform from various sensors,
An input circuit 9a having functions such as correcting the voltage level to a predetermined level and converting analog signal values into digital signal values.
, central processing circuit (hereinafter referred to as rcPUJ) 9b, C
storage means 9c for storing various calculation programs and calculation results executed by the PU 91), and the fuel injection valve 13
and an output circuit 9d that supplies a drive signal to the control valve 18.

The CPU 9b determines the operating state of the engine 1 based on input signals from the various sensors described above, and optimizes various characteristics such as fuel efficiency and acceleration characteristics according to the determined operating state. The fuel injection time and the like of the fuel injection valve 13 are calculated, and a drive signal based on the calculation result is supplied to the fuel injection valve 13 via the output circuit 9d.

In the above fuel injection time calculation, the basic fuel injection time, that is, the basic valve opening time Ti of the injection valve 13 is P BAT
The intake pipe absolute pressure PI based on the detection outputs of the C sensor 11 and the Ne sensor 14 is calculated from the 1゛i map (not shown) stored in the storage means 9C described above, according to the engine rotation speed Ne. Based on input signals from various sensors, the CPU 9b sets an oven loop control region, a feedback control region, etc. for boost pressure control according to the engine operating state, and determines whether or not the region is within those regions, according to a control program described below. At the same time, the opening degree of the movable vane 64, that is, the duty ratio D V a of the control valve 18, is calculated so that the optimum boost pressure can be obtained in the determined control region, and according to the calculated value. control valve 18
supplying a drive signal to actuate via the output circuit 9d,
Control valve 18 and further actuator 17 linked to turbocharger 4 are driven.

FIG. 4 shows a flowchart of a program for controlling the duty ratio Dvc of the control valve 18, that is, the boost pressure.

First, in step 401, I)v stored in the ECU 9
From the c4+ map, the reference value 1) ver+ of the duty ratio Dva is read out according to the throttle valve opening θTl+ and the engine speed Ne. 16 steps as e n+vt to OTIIV16 within the specified range, and engine speed Ne as 20 as Nl-NV2O within the specified range.
Each stage is provided, and the reference value Dvar+ is determined by interpolation calculation at points other than the grid points of the map. By setting the reference value DVOM using such a map, the duty ratio Dva of the control valve 18 can be adjusted to the engine 1
can be controlled in more detail depending on the operating state of the vehicle.

Next, it is determined whether the gear position of the transmission (hereinafter simply referred to as "gear position") is at the first speed (1, st) position (step 402). This determination is performed, for example, according to the subroutine shown in FIG.

That is, the predetermined speed Vt that is normally obtained when the vehicle speed ■ is in the first speed position
First, it is determined whether the value is smaller than step 601),
When V≧VL holds true, it is determined that the vehicle is not in the first speed position (step 602). On the other hand, in step 601, V (
When VL holds true, vehicle speed V is equal to engine speed N
Step 603) determines whether or not it is smaller than a predetermined value Vp corresponding to e. When V≧Vp holds, step 602 is executed, while when V(VF holds, it is determined that the gear is in the first speed position. (step 604)
.

FIG. 7 shows a table for determining the predetermined value Vp. That is, when the gear position is in the first gear position, the ratio between the engine speed Ne and the vehicle speed V is constant, so the reference value NPI~ of the engine speed Ne is adjusted to match this relationship. 'NP9 and vehicle speed V reference value VFI~Vp
e is set in advance as a table, and it is determined that the vehicle is in the first speed position when the vehicle speed V is smaller than a reference value Vp corresponding to the actual engine speed Ne. With this configuration, it is possible to determine whether or not the gear position is in the first gear position without using a gear position sensor, etc., not only when the transmission is a manual transmission, but also when the transmission is an automatic transmission. It can be done easily.

Returning to FIG. 4, the answer to step 402 is affirmative (Yes).
), that is, when the gear position is at the first speed position, the predetermined value DF is subtracted from the reference value DVGM of the duty ratio Dvc obtained in step 401 to obtain the base 7jll value D.
After resetting V(J (step 403), negative (No
), that is, when the gear position is at a position other than the first speed, the process directly proceeds to step 404. In this way, the reference value DVGM of the duty ratio Dvc is set to a value smaller by the predetermined value Dp when the gear position is at the first speed position than when the gear position is at a position other than the first speed.

In step 404, the Kvt^ stored in the ECU 9
The intake temperature correction coefficient KVT^ is read out from the map according to the engine speed Ne and the intake temperature T^.

FIG. 8 shows an example of this KVT^ map, where the engine speed Ne is Nv+ to NV similarly to the Dvcr+ map.
20 levels as 2O, intake temperature -17A is TAVl ~ TA
Eight stages are provided for V8, and the intake air temperature correction coefficient KVTA can be set more appropriately using such a map.

The intake air temperature correction coefficient KVT^ is a correction value for intake air temperature compensation (step 416.428), and others (7). At the main boost pressure ff1lJ#, the boost pressure, that is, the absolute pressure in the intake pipe PBATC, is controlled so that the torque is the target torque (see Fig. 19). The engine speed Ne
When the target supercharging pressure is changed according to the intake temperature T^, the opening degree adjustment of the movable vane 64 is also different, and therefore the duty ratio 'D of the control valve 18 is changed.
Since it is also necessary to change va, first, basically, the above-mentioned reference value DVGM, that is, the DVG11 value (
Step 401. , 403) by a predetermined coefficient value (value of the ratio) to obtain the necessary control mm.

For this reason, in step 404, each time this program is executed, the engine rotation speed N at the time is set as the coefficient for the multiplication correction described above.
The intake temperature correction coefficient 1 (VT^) is read out according to e and the intake temperature T^.

During oven valve control, for multiplication correction regarding the base value, in addition to the intake temperature correction coefficient Kvr^, the engine speed Ne and intake temperature T^ are also used as control amount correction. A later-described learning correction coefficient calculated for each area determined by and updated from time to time is also applied (see step 428), the details of which will be described later.

When the Kv7^ value is read in step 404, the amount of change (hereinafter simply referred to as "change rAJ") ΔP8^ in the intake pipe absolute pressure PIIATC is then calculated as the current value PB.
Calculate by the difference between Cn and the value PRATCn-3 three times before (step 405), , PH to this amount of change
^ is applied to set various constants for calculating the duty ratio Dva as described later, and thereby the increasing gradient of the supercharging pressure is controlled to a desired value.

Next, in step 406, it is determined whether or not the boost pressure is in a state that requires open loop control.7 This determination is performed according to the subroutine shown in FIG.

First, in step 001, the throttle valve opening degree On+ is
It is determined whether or not the opening degree is larger than a predetermined opening degree OPB indicating that the opening is in the fully open state, and if the answer is negative (NO), that is, (37
If H≦□pn is established and the throttle valve 7 is not in a substantially fully open state, it is determined that open loop control should be performed, and the process proceeds to step 916 to be described later. That is, the feedback control is executed only when the throttle valve 7 is substantially fully open.

If the answer to step 901 is Yes, that is, the throttle valve is almost fully open, step 902) determines whether feedback (F/B) control was performed in the previous loop. is 17 constant (Ye
At step s), it is determined that feedback control should be continued, and the program is terminated at step 903).

If the answer to step 902 is negative (No), that is, open loop control was performed last time, step f is performed to determine whether or not the gear position is at the first speed position.
304), when in a position other than 1st gear, E C
From the PBASD table stored in U9, a first subtraction value APIIASD for a position other than the first speed is determined according to the amount of change ΔpB^ (step 905), and the process proceeds to step 907, which will be described later. Figure 1o shows an example of this ΔPBASD table, in which two reference values ΔPH^1 and ΔPH^2 (〉△Ps^1) are set for the amount of change APII^, and the larger the ΔPs^ value, the more The larger the increase gradient of supply pressure is, the larger the first subtraction value APIIAsp becomes.
ΔPB for less than B^2 and △Ps^2 or more, respectively.
A13 to ΔPBASDI are set.

When the answer to the step 904 is 1\constant (Yes), that is, the gear position 11 is at the first speed position, the first subtraction value ΔP8^SD is changed to the predetermined value ΔP[1AS11 for the first speed position.
If the value is set to F (step 906), the process proceeds to step 907.

The predetermined value ΔI) RASDF is set to a value larger than the APBASD value for positions other than the first speed, which is obtained in step 905.

Next, in step 907, the intake pipe absolute pressure! '[B
ATC sets its target value P8^TRO and the step 00.
5 or the first subtraction value obtained in 906 APIIAS
D (PBATko-ΔPBASD) (hereinafter [
It is determined whether the pressure is greater than the "minimum opening control release pressure".

The target value P aArgc of the intake pipe absolute pressure is set according to the engine speed Ne and the intake air temperature T^ in the control program shown in FIG. 4 (see step 408), as will be described later. 20), and furthermore, in this example, it is also set according to the gear position (step 410).
).

If the answer to step 907 is negative (No), that is, the intake pipe absolute pressure PIIATC is the minimum opening control release pressure (PIIA
rga−ΔPBASD), both the proportional control term DVP and the integral control term Dvi, which will be described later and are applied to feedback control, are set to 0.0 in step 908.
909), and then the duty ratio Dvc is set to 100%, that is, the movable vane 64 is set to the minimum opening degree (step 910). That is, P IIATC≦(P BAT
When RG ・−△PIIAsD) is established, the minimum opening control of the movable vane 64 is executed (from 0 to t in FIG. Maximize the rising slope 1. The responsiveness of the boost pressure control can be improved by controlling the boost pressure to increase the boost pressure to near the desired pressure value.

Next, the feedback control delay timer t and FBDLY are reset (step 911), and the process proceeds to step 418 in FIG. 4, where a drive signal based on the duty ratio Dvc is output to the control valve 18, and the control program shown in FIG. end.

Returning to the subroutine in FIG. 9, step 90 of nIj is affirmative (Yes), that is, the absolute pressure in the intake pipe P II
ATc is the minimum opening control release pressure (PBATko - ΔPBA
SD), it is determined whether the gear position is in the first gear position or not (step 912), and if it is in a position other than the first gear, the APIIA stored in the ECU 9 is
From the P11 table, the first
Find the second subtraction value ΔP BAFB for positions other than speed (
Step 913), proceed to step 91.5 described below.

FIG. 11 shows an example of this AP RAFB table,
Just like in Figure 10, the base 7F! for the amount of change APR^!
Value ΔPH^1. ΔI)B^2 is set, APIIA1
less than, △PR^1 or more and less than ΔPBAs, and △l) B^
ΔPIlAr[I3~ΔPR
AFB1(ΔPBAPB3<ΔPBApa2(ΔPBA
FBI) is set.

When the answer to step 912 is 11 (Yes), that is, the gear position [l is at the first speed position, the second subtraction value ΔP BApn is set to a predetermined value for the first speed position PBAFRF
(step 914), the process proceeds to step 915.

The predetermined value P BAPBF is set to a value larger than the AP BAPBF value for positions other than the first speed, which is obtained in step 913.

Next, in step 915, the intake pipe absolute pressure PBAr
c is the target value P BATRG and the step 913
The difference from the second subtraction value ΔPBAFB found in step 914 (P BATIIC - ΔPBAFB) (hereinafter referred to as “
It is determined whether the pressure is greater than the "fiber pack control start pressure" (step 915). This answer is negative (No
), that is, when the intake pipe absolute pressure I]BATC is less than the feedback control start pressure (PRATRG-ΔPBAFB), it is determined that open loop control should be performed, and the process proceeds to step 916 and subsequent steps. That is, (P BATII
C-ΔPBASD)<PR^DingC≦(PI]^TR[;
-ΔP8AFB) is established, open loop control is executed (between t and tB in FIG. 18).

In step 916, the tPBDLY timer is reset as in step 911, and then it is determined whether the gear position is at the first speed position (step 917).
. When the answer is negative (NO), the DT child table stored in the ECU 9, etc., according to the amount of change ΔPa^,
A subtraction term DT for positions other than speed Ⅰ, which will be described later, is found (step 918), which is applied to oven loop control, and the process proceeds to step 921, which will be described later.

FIG. 12 shows an example of this Dr child table, and in exactly the same way as in FIG. 10, the basic value v1 value ΔPB^
1. ΔPB^2 is set, less than ΔPB^1, Δl'B
DTI to DT3 (D1<DT2<DT
3) is set.

When the answer to step 917 is affirmative (Yes), that is, when the gear position is at the first speed position, the first speed is changed according to the i:1 change amount △PB^ from the DPT table stored in the ECU 9. Find the subtraction term DFT for position (step 91
9). FIG. 13 shows an example of this DFT table, in which two reference values ΔPBAFI and Δ
PBAF2 (>ΔPBAPI) is set and ΔPEA
less than rt, △PBAFI or more but less than ΔPBAF2 and ΔP
For t+Ap2 or more, Dp 1~D FT3 respectively
(DFTI<DFT2<DFT3) is set.

Next, the subtraction term Dτ is set to the subtracted DFT value (step 920), and in step 921 it is determined that open loop control should be executed, and the program is ended.

Thus, when the throttle valve 7 is almost fully opened, the above-described minimum opening control that forcibly sets the duty ratio Dvc to 100% is first executed, and when this control is released, open loop control is executed. become.

Specifically, the open loop control is executed according to the calculation of the duty ratio Dvc, the limit check, and the output processing in steps 425 to 249.418, which will be described later in FIG.
8. PBAT changes as shown in Figure 6.
Although the degree of "-H" is temporarily relaxed at time L^, the C value still shows an upward trend.

In step 915, in this process, the magnitude of the supercharging pressure, that is, in this embodiment, the intake pipe absolute pressure Pa
The value is monitored, and the answer to step 915 is affirmative (Y
es), that is, when the intake pipe absolute pressure PB8C exceeds the feedback control start pressure (PBATRG-△PBApB) once, the FBDLY timer is reset in step 911 or 91G (the example in FIG. 18). Then, it is determined whether a predetermined time tFBDLY7'l< has elapsed (step 922), with the reset time at step 916 as the measurement starting point. This answer is negative (No)
When this is the case, the process proceeds to step 917, where open loop control is performed, while when the answer is yes, it is determined that feedback control should be performed, and the process proceeds to step 923.7 In this way, the intake pipe absolute pressure PB8 is changed to C.
is the feedback control starting pressure (P BATI!O−ΔP
13^FB) twice, rather than performing feedback control immediately, the feedback control is performed for a predetermined period of time t FB from this time.
Open loop control is executed until DLY has elapsed (between tB and t and c in FIG. 18), and then feedback control is executed for the first time (after C in the figure).

As a result of the above, the intake pipe absolute pressure P BATC is set to the predetermined value (P
BATI! G-ΔP BApB) or more and a predetermined period of time has elapsed, it is determined to be in a steady state and feedback control is started, so stable boost pressure control performance is maintained even when transitioning from open-loop control to feedback control. can be obtained.

When entering the feedback control region, the actual intake pipe absolute pressure P BATC is determined through steps 923 and 903 and step 407 in FIG.
Feedback control is executed to determine a control amount and perform supercharging pressure control so that the PBATC value becomes the target value PDATRO according to the deviation between the PBATC value and the target value PBAtgc.

That is, in step 923, the initial value of the integral control term Dvi is calculated according to the following equation (1).

Dvi=Kvr^XDvar+X (KvRppij-
1)-(1) Here, Kvkppij is a learning correction coefficient (learning value) calculated at the time of feedback control as described later according to the program shown in FIG.

As mentioned before, the learning correction coefficient Kvappij is a learning value for intake temperature correction obtained by performing learning control during feedback control for each region determined by the intake temperature ゛■゛^ and the engine rotation speed Ne. , 1U, Step 92
In step 3, since the initial value is calculated using this, the setting of the control meter for the control valve 18 at the time of transition to feedback control is appropriately performed. That is, when entering the feedback control region, the initial value of the integral control term Dvi is set at step 9.
23, but in that case, as shown in equation (1) above, the learning h+i positive coefficient Kvt! by the learning control up to that point, that is, the conventional feedback control. Since the initial value is calculated by applying ppij, the control amount initially applied in the control at the time of transfer to the feedback control area can also be very similar.

Next, the process proceeds to step 903, where it is determined that feedback control should be performed and the program is terminated.

Returning to the program in FIG. 4, in step 407 following step 406, it is determined whether or not the sub-routine in FIG. 9 executed in step 406 has determined that open loop control should be performed.

If the answer is negative (No), that is, it is determined that feedback control should be performed, the intake pipe absolute pressure is ]1;j memory-
1 standard price Pn^1'gG is read (step 408).

FIG. 14 shows an example of this Pncna map. 1 Just like the above KVT^ map, the reference values Nv1 to NV2O of the engine speed Ne and the reference value TAVI of the intake temperature T^
~ TAV8 is set, and the P BATi! By applying the readout value of the G map as the target value for boost pressure control during feedback control (steady state), appropriate variable control of the target pressure is performed according to the engine speed Ne and intake temperature .

Regarding the change of the PBArgc value according to each parameter, specifically, the boost pressure is low at low rotation speed and low intake temperature, and the boost pressure decreases to large rlr as the rotation speed increases. Can be set to

In the following, regarding the setting and change control of the target pressure according to changes in the intake air temperature T^ and the engine speed Ne, the intake air temperature (the temperature detected by the intake air temperature sensor 12, which is different from the outside air temperature, and is the temperature inside the intake pipe) When performing feedback control by setting the target pressure so that the engine output remains the same regardless of the temperature of the intake air just before it is sucked into the engine via the turbocharger 4 and further through the intercooler 6. The description will be made with reference to FIG. 19, which shows the target torque characteristics, and FIG. 20, which shows an example of characteristics regarding the set pressure for obtaining the target torque.

The control according to this program aims to achieve a smooth acceleration feeling for a practical vehicle even with a highly supercharged engine, and to achieve an optimal supercharging control system that can respond to changes in environmental conditions. - As a link, the target supercharging pressure is set based on the intake temperature T^ and the engine speed Ne, and it is possible to prevent excessive rise in exhaust temperature and engine damage due to over-output at low intake temperatures, or to prevent engine damage at high intake temperatures. Executes control to prevent insufficient output, etc. In other words, the intake temperature
^ is an environmental factor that has a large effect on direct boost pressure, and its changes affect charging efficiency, so if the intake temperature ゛r^ changes, even if the intake temperature If the boost pressure characteristics were set to be the same, the output would be excessive at low intake temperatures, and the output would be insufficient at high intake temperatures. Therefore, the boost pressure is also affected by the engine speed Ne (even if the intake air temperature is the same, it differs depending on the engine speed).

More specifically, as shown in FIG.
When it is desired to always obtain substantially the same torque (same horsepower) in the range from m to high rotational speed range), it is difficult to obtain such a target torque characteristic unless the target pressure is controlled to change. That is, as described above, the turbocharger 4 compresses (pressurizes) intake air and sends it to the engine according to the rotational speed of the turbine wheel 61 that corresponds to the flow rate of exhaust gas. Since the air density is different in summer when the outside temperature is high and in winter when the outside temperature is low, this causes a difference in the output that can be obtained.

In other words, even if the same volume of air is compressed and sent, the same torque cannot be obtained;
For example, if the outside temperature is normal (for example, 20 to 30 degrees Celsius), when it is compressed, it will reach a fairly high temperature (for example, about 80 degrees Celsius) (even with the intercooler 6, the temperature will still be low). In that case, the desired target torque cannot be obtained unless the pressure is supercharged to a larger value.Conversely, if the outside temperature is lower than the normal value (e.g. O ℃), in order to compress the air at a lower outside temperature and obtain the same target l-lux as the double series, the pressure must be lower than that at the high outside temperature. It is necessary to lower it by a predetermined value.

From the above point of view, in order to avoid the inconvenience of a difference in output and to always obtain almost the same engine output as shown in Fig. 19, it is necessary to In order to change the supercharging rate depending on the temperature at the time of pumping, the target absolute pressure in the intake pipe
An example of this is set to the characteristics shown in FIG. The information is mapped into an O map and stored in advance.

The set pressure characteristics for obtaining the target torque in FIG. 20 are exemplified by engine speed Ne and intake temperature T^ (regarding intake temperature T^, a case where predetermined temperatures TAI, T^2°TA3 are exemplified, These relationships are T^]〉T^2>T^3
In the illustrated example, the amount of change in target boost pressure due to changes in intake temperature is generally determined by changing the amount of change in target boost pressure due to changes in intake temperature in the high rotation speed region or normal rotation speed region. It is set so that it is larger than the rotation speed region. That is, in these regular use and high rotation ranges where drivability is especially important, the target pressure is set to be smaller than when the intake temperature is low, in order to correct the influence caused by changes in intake temperature.

Regarding the set pressure characteristics depending on the intake temperature T^ and the engine rotation speed Ne, depending on the engine to which it is applied, for example, in Fig. 20, on the lower rotation speed side, each T^1 ~
In the case of TAXI, the width between the characteristics becomes narrower, and the width becomes wider on the high rotation speed side, etc.
As shown in the figure, it is possible to appropriately determine the torque characteristic so that it becomes approximately flat within the necessary engine rotational speed Ne range.

As described above, P8ATRG is used to control the change in the target value of boost pressure during feedback control based on intake temperature T^ and engine speed Ne, that is, during control in a steady state.
If the values are mapped and read and applied during the control, the target value P can be adjusted using the above r'aArgc map, even compared to the case where the target value is corrected only by the intake air temperature in lit.
It is possible to set B^DRG even more appropriately, and the target pressure can also be controlled by changing the engine rotation speed Ne, and the torque that reaches the rotation speed that is considered to be the best initially set, that is, the necessary torque (target torque) This allows more appropriate control to be achieved.

As described above, this supercharging pressure control controls the intake pipe absolute pressure PBATC so that the torque reaches the target torque, regardless of changes in the intake air temperature and in appropriate response to the engine speed Ne. As mentioned above, the duty ratio basic amount corresponding to the engine speed Ne, that is, the reference value [)
In addition to van (step 401), it has a correction term based on intake temperature TA, that is, KVTA value (step 404), and also has a target value P nAr1:c map for making torque according to engine speed Ne and intake temperature T^. When the feedback control is performed, step 408 is performed.
In this step, a search for the map is first performed.

However, when the P BATRC value is read, next...
In step 409, it is determined whether or not the gear position is at the 951st speed position, and if the gear position is at the 1st speed position, step 409
From the target value PBATRG obtained in step 08, the predetermined value P IIA
TI! After subtracting OF (step 410) and resetting the target value P_BATRC, the process directly proceeds to step 411 when the gear is in a position other than the first speed. in this way,
The target value P IIATRO is set to a smaller value by a predetermined value P BArtap when the gear position is at the first speed position than when the gear position is at a position other than the first speed.

In this step 411, step 408 or 41
Deviation ΔP B (= P aArRa
-P IIATC) is calculated, and then the absolute value 1ΔPH1 of the deviation ΔPa is set to a predetermined value Gai (for example, 20 ml
1g) or more (step old 2).

This predetermined value GBi is a dead zone defining pressure during feedback control.

The answer to step 412 is Yes, that is, 1ΔP.
When BI≧Gsi holds true, VP and Kvi are read out from the KVP table and Kvi table stored in the ECU 9 as constants of the proportional control term DVP and integral control term Dvi, respectively, according to the engine speed Ne ( Step 413). Figures 15 and 16 show this K.
FIG. 6 is a diagram showing an example of a vr table and a Kvi table, respectively. That is, in the KVP table, two reference values NKVPI and NKV are set for the engine speed Ne.
P2 (>NKVPI) is set, and the constant KVP is N
Less than xvrt, NKVPI or more but less than NKVP2, and NK
For VF2 and above, Kvrt- and vr3(
Kvrl<Kvr2<Kvr3), and in the Kvi table, the engine speed Ne
Two reference values Ncvix and NKviz (>
NKvix) is set, and the constant Kvi is Ns;vit
less than, N Kvi l or more Jz NKvi less than 2 and Nr,
Kvi 1 to Kv for viz -1=, respectively
It is set to ia(Kvi:+(Kvix(Kvi 2)).

Next, the proportional control term 1) VP is changed to the constant I<
, set the integral control term Dvi to the product KVP·ΔPH of the constant Kvi obtained above and the deviation ΔPB.Step 4I4)
and the integral control term Dvi calculated up to the previous time (=
1) Set vi + Kvi ・ΔPB) (step 4!5).

Next, the proportional and integral control terms DVP and Dvi set in the above are applied to calculate the duty ratio Dva during feedback control according to the following equation (2) (step 416).

Dvc=DvciXKvr^+DvP+Dvi −(
2) Next, perform a limit check on the calculated duty ratio 1 VG to maintain the duty ratio pva within a predetermined range (step 417), and further perform drive based on the duty ratio Dva. A signal is output to the control valve 18 (step 4I8) and this program is ended.

The Dva value calculated by the above formula (2) is calculated in step 417.
If it is within the range of the check, this is the final output, and the first term (D
vcr+X KVT^), the basic control amount corrected by the intake air temperature]
The value obtained by adding the calculated correction ff1 (Dvr+Dvi) in 4.415 is output as the control amount. Thus, the actual boost pressure, i.e., the P BATC value and 1);
The control amount during feedback control is determined according to the deviation from the target value PIIATRG based on the gc map, and based on this, the boost pressure is adjusted to the engine speed Ne and intake temperature T^
(Fig. 20), and as a result, the engine output becomes 19.
As shown in the figure, almost the same characteristics can always be obtained,
Engine damage and insufficient output can be reliably prevented.

Under such control, when certain conditions are met, learning control is executed as described below.

That is, during feedback control, each time the step 412 is executed, the difference 1ΔPB1 between the intake pipe absolute pressure P BATC, which is the actual boost pressure value, with respect to the target value P BATRO at the time is monitored, and as a result, the step 412 is monitored. 412
If the answer is negative (No), that is, 1△PBi<G++i holds true, and therefore the target value P BATRG and the actual intake pipe absolute pressure P IIATc almost match, the proportional control term DVP is set to the value 0.0. , the integral control term Dvi is set to its previous value Dvi (step 41!13
．． 420).

Next, it is determined whether the gear position is at the first speed position (step 421), and if the answer is affirmative (Y'es),
That is, when the gear position is at the first speed position, steps 422 to 424 described below are skipped and step 416 is performed.
Proceed directly to the following, and if the answer is negative (No), i.e.
When the position is other than the first speed, the coefficient KVR is calculated according to the following equation (3) (step 422).

KvRKvr^XDvcn+Dvi ,, (3)
-vT^×vG This coefficient KVR is tJ representing the deviation in boost pressure control due to variations due to mass production or changes over time.

In other words, the value (KVTAX DVG+1) obtained by multiplying the intake temperature correction coefficient KVT^ and the reference value DVI is determined by the above (2).
As mentioned in the explanation of the formula, it is a basic control amount corrected by the intake temperature T^, and if it is assumed that there are no individual differences at the time of manufacture of the components used or changes over time, then the intake temperature T This is supposed to be the correct value for the basic f control fee corrected for the intake temperature at From this point of view, feedback control sequentially detects the deviation between the target boost pressure and the actual boost pressure, and adjusts the deviation so that the deviation becomes zero (specifically, as described in step 6 above). In some cases, the integral control term Dvi in the equation (3), that is, the previous value Dvi set in step 420, is Since this deviation is reflected, the above-mentioned KvI! as a value, i.e. (KV
The sum of the TA+DVGM) value and the Dvi value is the corresponding (KV
It is calculated by dividing the value by the TA+DVOM) value.

Next, the KVR value thus calculated is subjected to a predetermined weighting process, and control is executed to rewrite the learning value for intake temperature correction using the engine speed No. and intake temperature 1゛^ region division force formula.

That is, next, the learning correction coefficient Kv is calculated using the coefficient KVR.
gtpi,i is calculated according to the following equation (4). (Step 423).

Kvxppij-hourX Kvg +ArX KvREr
ij...(4) Here, the KvgEpij value in the second term on the right side is the learning correction coefficient obtained up to the previous time, and is read from the KVREP map described later according to the engine rotation speed Ne and the intake temperature T^. . Further, A is a constant, and Cvu:p is a variable that is experimentally set to an appropriate value in addition to 1-A.

K for KvREpij by the value of variable CVREP
Since the ratio of the VR value changes, the optimal Kvxspij can be achieved by setting this CVREF value to an appropriate value in the range from 1 to 1 above depending on the use of the target boost pressure control device, engine, etc. can be obtained.

Next, the learning correction coefficients KvtP, pij calculated above are
is provided in the backup RAM in the E CtJ Q (a memory that retains memory even when the ignition switch is turned off, and such backup memory may be, for example, a non-volatile memory or other appropriate memory). Remember in KVREF map step 42/
I) Execute the steps from step 416 onwards and terminate this program. FIG. 17 shows an example of this KVREF map. That is, 1 (VRP, F map is miki Kv ding^ map (Fig. 8) and PBATl!G map (Fig. 14)
Similarly, the engine speed Ne and the intake air temperature 'r^ are divided into a plurality of regions, and the KvgEPjj value is calculated and stored for each region to which the Ne value and the T^ value apply.

As mentioned above, when performing learning control during feedback control and using the learned value as a control amount correction during open loop control, it is not possible to simply divide the intake temperature correction learned value for each intake temperature range and have it. Instead, in addition to the intake temperature T^, the engine speed Ne is also used as a factor, and the intake temperature correction value is calculated by learning control for each region determined by the intake temperature T^ and the engine speed Ne. Find the value of K v i! of each region. Rewriting and storing the E p value (steps 422, 423), that is, in updating the KVREP value, the step 42
Every 2 executions, the engine rotation speed N at the time of the current loop execution
Kvu for the area including the e value and the intake air temperature TA value
:The pij value will be rewritten one after another.

The KVREP map used in steps 422 and 423 uses the Ne value and T^ value as factors in the same way as the KVTA map and PBArga map described above, but more specifically, if λ, the Ne value and r^ value are used. The defined area is considered as one unit, and Kvgl:p is applied to this area.
Preferably, the ij values are calculated and stored, ie, updated.

This is from the following points of view.

In the case of the KVT^ map and the PBAyRc map, as in the case of the DVGM map described above, when the intake temperature T^ and the engine speed Ne are values other than the map grid points,
By interpolation calculation (4-point internal pressing method), the corresponding Kv ding^ij value, P BArt! In the case of the Kv*pp map, the cij value is determined, and in the case of the Kv*pp map, learning is performed for each region, that is, between temperature values consisting of intake air temperature values, and between rotation values consisting of engine rotation values consisting of a row. Specifically, as shown in FIG. 17, the intake temperature 'r^ is T^■1~T Av,j~
Divided by TAV8 and engine speed Ne
is divided by Nv+~Nvi~NV2O, the r^ value and Ne value at the time of execution of step 422 are, for example, values near TAvj (, j=1.~,8). Nvi (i = 1, ~, 20
), that is, if the operating state is exactly such, the learned value (Kvgppij in this example) is rewritten only for the corresponding region.

In this way, by performing learning in area units and rewriting only the relevant area in area units, the T^ value, N
Even when the e value takes an intermediate value between adjacent reference values other than the reference value mentioned above, there is no need to divide the calculation into four points as in the Kvr^ map and P BATRO map, and learning control and its In controlling the rewriting of learned values, the processing time can be reduced. When applying a method that calculates a negative value by calculating a target value, etc., the value must be returned to its original value after learning. Therefore, in the case of an intermediate value, conversely it is divided into four points. (This requires additional processing time.)

Therefore, as described above, during learning control, learning control is performed for both the intake temperature T^ and the engine speed, and the learning correction coefficient is rewritten for each corresponding region.

Thus, if the above rewriting process is performed, the rewritten Kvl! ppij(+l'() is an appropriate value as a correction amount for each deviation of the control amount with respect to the corresponding intake temperature 'r^ and engine speed Ne, so as mentioned above, depending on the intake temperature T^ In order to set the optimum boost pressure and prevent engine damage and insufficient output, the intake temperature 'r
Even if a boost pressure control method is adopted in which the target value of boost pressure is set using ^ and the engine speed Ne (step 408, etc.), the value learned at a different 11 standard pressure may be incorrectly rewritten. It is also possible to avoid situations where Kvlppij
V according to the Ne value and 'r^ value at that time from the map.
By reading the vtppij value and applying it, it is possible to modify the control amount so that the control a1ffi becomes a more correct value when in the oven loop control state.As a result, when updating the learned value, the learned value for intake temperature correction It is possible to obtain a more appropriate learning value than when dividing and rewriting for each intake temperature, and it is possible to set the boost pressure during open-loop control with higher accuracy.

That is, when proceeding from step 406 in FIG. 4 to determination step 407, the answer to step 407 is
7 constant (Yes), that is, when it is determined by the subroutine of FIG. 9 that open loop control should be performed, the learning correction coefficient Kvg[1pij is Read step 425), then proportional system four terms Dv
Both r and integral control term Dvi are set to the value 0.0 (
Steps 426, 427).

Next, the duty ratio Dv during open loop control
a is calculated according to the following equation (5) (step 428).

Dvc=Kv ding^XKvI! Epijx (Dvar
+-DT) -(5) Here, DT is the subtraction term set in step 918 or 920 of the subroutine of FIG.

Next, perform a limit check on the duty ratio Dva calculated above, and set the Dva value to 0% or more and 100%, for example.
The value is maintained at the following value (step 429), jγj row (step 418), and the program ends.

As described above, the learning correction coefficient Kvgppij is determined based on the engine speed Ne and the intake temperature T during feedback control.
^ is calculated and stored for each applicable area, and T^ and Ne values are applied to the calculation of the duty ratio Dva for each applicable area during oven loop control. Therefore, learning control is performed during feedback control,
When using the learned value to correct the control amount during open-loop control, it can accurately correct the deviation in boost pressure control described above depending on the intake temperature T^ and engine speed Ne, which have a large effect on boost pressure. This allows for appropriate supercharging pressure control during oven loop control.

In this embodiment, the supercharging pressure is detected using the intake pipe absolute pressure PBATC sensor downstream of the throttle valve, but a dedicated pressure sensor is installed in the intake pipe upstream of the throttle valve and downstream of the turbocharger. A supercharging pressure detection sensor may be provided to detect the supercharging pressure.

Furthermore, in this embodiment, as a supercharger capable of controlling boost pressure,
Although the description has been given using an example of a turbocharger in which the capacity is changed by operating movable vanes, the present invention is not limited to this, and can also be applied to wastegate type and supercharging pressure relief type turbochargers. Furthermore, the present invention may also be applied to control of boost pressure in a so-called supercharger driven by the output power of an engine.

(Effects of the Invention) As described above, according to the present invention, in the method for controlling the boost pressure of an internal combustion engine, which controls the boost pressure by controlling the control amount of the boost pressure control means, the target boost pressure is is set according to the intake air temperature and the engine speed, and the control amount is feedback-controlled according to the deviation between the actual boost pressure and the target boost pressure when the boost pressure is in a steady state, and the When the boost pressure is in a transient state, open-loop control is performed based on the learned value calculated during the feedback control, and the learned value is set in a plurality of regions determined by the intake air temperature and engine rotation speed during the feedback control. Since it is updated to the value obtained by learning each time, even if the target value of boost pressure is changed depending on the intake temperature and engine speed, the learning control during feedback control always To be able to obtain an appropriate learned value corresponding to the intake temperature and engine speed, and therefore to prevent values learned and obtained at different target pressures from being applied during open loop control. As a result, the control amount can be set accurately even during open-loop control, and boost pressure control can be performed more appropriately than when simply obtaining a learned value by dividing it for each intake temperature.

[Brief explanation of the drawing]

The drawings show one embodiment of the present invention, and FIG. 1 is an overall configuration diagram of a control device for an internal combustion engine equipped with a turbocharger to which the control method of the present invention is applied, FIG. 2 is a longitudinal sectional view of the turbocharger, and FIG. Figure 3 is an arrow view of the turbine casing side viewed from ■-■ in Figure 2, and Figure 4 is the duty ratio D of the control valve.
FIG. 5 is a diagram showing a map of the reference value DVOM of the duty ratio Dvo, FIG. 6 is a flowchart of a subroutine for determining the gear position of the transmission, and FIG. 7 is a subroutine of FIG. 6. FIG. 8 is a diagram showing a map of the intake temperature correction coefficient KVT^, and FIG. 9 is a flowchart of the open loop control region determination subroutine executed in step 406 of FIG. 4. , FIG. 10 shows a table of the first subtraction value APIIASD for positions other than the first speed, and FIG. 11 shows the second subtraction value ΔP for positions other than the first speed.
FIG. 12 is a diagram showing a table of subtraction term DT for positions other than 1st speed, FIG. 13 is a diagram showing a table of subtraction term DFT for 1st speed position, and FIG. 14 is a diagram showing a table of IIAPR. is a diagram showing a map of the target value P [1 ATRO of boost pressure, Figure 15 is a diagram showing a table of constant KVP of proportional control term VP, and Figure 16 is a diagram showing a table of constant KVP of integral control term Dvi.
A diagram showing a table of i, FIG. 17 is a learning correction coefficient KVR
A diagram showing a map of EF, Figure 18 shows the absolute pressure P in the intake pipe.
FIG. 19 is a diagram showing the relationship between BATC and boost pressure control, FIG. 19 is a diagram showing an example of target torque characteristics used to explain the present invention, and FIG. 20 is a diagram showing an example of set pressure characteristics for obtaining the target torque. It is a diagram. l... Internal combustion engine, 4... Turbocharger, 9
...Electronic control unit, 11...Intake pipe absolute pressure (P BATC) sensor, 12...Intake temperature (T
^) Sensor, 14... Engine speed (Ne) sensor, 17... Actuator, 18... Solenoid control valve (
supercharging pressure control means), Pe^TC...absolute pressure in the intake pipe (
Boost pressure), Dvc...Duty ratio (control amount), P
BATRC...Intake pipe absolute pressure P BATC target value (target boost pressure), Kvt! ppij...Learning correction coefficient (learning value). Applicant Honda Motor Co., Ltd.

Claims (1)

[Claims] 1. In a method for controlling the boost pressure of an internal combustion engine in which the boost pressure is controlled by controlling the control amount of the boost pressure control means, the target boost pressure is set according to the intake air temperature and the engine speed, and the The control amount is feedback-controlled according to the deviation between the actual boost pressure and the target boost pressure when the boost pressure is in a steady state, and calculated during the feedback control when the boost pressure is in a transient state. Open loop control is performed based on the learned value, and the learned value is updated to a value obtained by learning for each of a plurality of regions determined by the intake air temperature and engine speed during the feedback control. A method for controlling the boost pressure of an internal combustion engine.