JP2004116371A - Engine controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve controllability by surely judging whether air-fuel ratio learning is sufficiently advanced or not and allowing the air-fuel ratio learning and canister purge controlling to be compatible. <P>SOLUTION: When a backup RAM is not cleared and an atmospheric pressure range is not changed, whether a mean value LAMBDAAVE of an air-fuel ratio feed-back correction coefficient falls within a set range to a reference value or not is checked (S505). If the value LAMBDAAVE falls within the set range, it is judged that the present learning range learning is completed (S506), and whether the learning completion judging range number is a set range number or more or not is checked (S507). As a result, if the number of the learning completion judging range is a set number or more, a purge execution allowing flag FLG is set (S508), and execution of canister purging is allowed. Thus, whether the air-fuel ratio learning is sufficiently advanced or not is surely judged, and the air-fuel ratio learning control and the canister purge control are allowed to be compatible, and hence the controllability can be improved. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an engine control device that improves controllability by achieving both air-fuel ratio learning control and canister purge control.
[0002]
[Prior art]
As is well known, in an air-fuel ratio control system of an engine, a variation in tolerance with respect to an engine specification, that is, a variation in the production of an intake air measurement system such as an intake air amount sensor or a fuel supply system such as an injector, or a change with time. In order to quickly correct the deviation of the air-fuel ratio, air-fuel ratio learning control is incorporated in feedback control by an air-fuel ratio sensor such as an O2 sensor, so that the state of the target air-fuel ratio is always maintained even when the operating state changes significantly. Like that.
[0003]
In the air-fuel ratio learning, if the canister purge control for purging the fuel vapor from the canister storing the fuel vapor in the fuel tank to the intake system is executed, the air-fuel ratio change due to the canister purge is erroneously learned. For this reason, generally, the air-fuel ratio learning is not performed during the execution of the canister purge. However, when the purge execution frequency is high, the air-fuel ratio learning frequency decreases, and the engine tolerance due to the learning cannot be absorbed. In this state, the air-fuel ratio is corrected when the canister purge is executed, and the controllability deteriorates.
[0004]
To cope with this, Japanese Patent Laid-Open Publication No. Hei 6-101581 discloses an air-fuel ratio learning control by counting the number of times the air-fuel ratio learning value is updated and performing a purge when the number of times the learning value is updated is equal to or greater than a predetermined value. There is disclosed a technique for achieving both the purge control and the canister purge control (see Patent Document 1). In Japanese Patent Application Laid-Open No. 11-166455, a counter for counting the number of times of execution of learning is set in a plurality of learning areas of an air-fuel ratio learning value storage map, and the number of times of learning is set in all the learning areas where this counter is set. A technique is disclosed in which the purge is limited (prohibited) when the number is less than the number of times, thereby achieving both the air-fuel ratio learning control and the canister purge control (see Patent Document 2).
[0005]
[Patent Document 1]
JP-A-6-101581
[0006]
[Patent Document 2]
JP-A-11-166455
[0007]
[Problems to be solved by the invention]
However, in the technique of Patent Document 1, the canister purge is executed only when the number of times the learning value is updated reaches a predetermined number in a part of the area, and the canister purge is performed in a state where the learning is not sufficiently advanced in the entire area. May be executed. Also, in the technique of Patent Document 2, similarly, whether or not canister purging is performed is determined only by the number of times of learning, so that the traveling environment changes from lowland to highland or from highland to lowland, and the driving conditions change. Thus, even if the learning value is expected to fluctuate, the canister purge may be executed.
[0008]
That is, in the prior arts disclosed in Patent Literature 1 and Patent Literature 2, execution / non-execution of the canister purge is determined only by the number of times of learning, and learning is performed for the entire operation region and various operation conditions. The determination as to whether or not to execute the canister purge is made by reliably grasping whether or not the progress is sufficient. In other words, simply determining whether to execute the canister purging based on the number of times of learning can only be a rough estimation such as setting the number of times assuming the maximum variation with respect to the variation in the air-fuel ratio due to the variation in engine tolerance. The original function of air-fuel ratio learning cannot be reflected in the air-fuel ratio control at the time of execution.
[0009]
The present invention has been made in view of the above circumstances, to reliably determine whether or not air-fuel ratio learning has sufficiently progressed, and to improve controllability by achieving both air-fuel ratio learning control and canister purge control. It is an object of the present invention to provide an engine control device capable of performing the following.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, an invention according to claim 1 provides an air-fuel ratio for learning a deviation from a reference value of an air-fuel ratio feedback correction coefficient set based on an output of an air-fuel ratio sensor interposed in an exhaust system of an engine. An engine control device comprising: a learning means; and a purge control means for controlling an amount of evaporative fuel purged from a canister for storing evaporative fuel in a fuel tank to an intake system of the engine. The air-fuel ratio learning completion determining means for determining whether the air-fuel ratio learning has been completed, and the purge control means when the number of regions determined to be learning completed by the air-fuel ratio learning completion determining means is equal to or greater than a set number. Purge execution permission means for permitting execution of purge of evaporated fuel from the canister to the intake system.
[0011]
According to a second aspect of the present invention, in the first aspect of the invention, the air-fuel ratio learning completion determining means performs the air-fuel ratio feedback without shifting the atmospheric pressure between a plurality of previously divided atmospheric pressure regions in the current learning region. When the average value of the correction coefficient is within the set range, it is determined that the air-fuel ratio learning is completed.
[0012]
That is, the invention according to claim 1 determines whether or not the learning is completed for each learning region of the air-fuel ratio learning, and when the number of regions determined to be learning completed is equal to or greater than a set number, the canister to the intake system is determined. By permitting the execution of the purge of evaporative fuel, it is possible to reliably determine whether the air-fuel ratio learning has sufficiently progressed in various operating regions and operating conditions, and to achieve both air-fuel ratio learning control and canister purge control. To improve controllability.
[0013]
At this time, it is determined that the air-fuel ratio learning is completed when the average value of the air-fuel ratio feedback correction coefficient is within the set range without shifting the atmospheric pressure between a plurality of previously divided atmospheric pressure regions in the current learning region. It is desirable.
[0014]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings. 1 to 10 relate to an embodiment of the present invention. FIG. 1 is an overall view of an engine system, FIG. 2 is a circuit configuration diagram of an electronic control system, FIG. 3 is a flowchart of a fuel injection control routine, and FIG. FIG. 5 is a flowchart of a fuel-ratio feedback correction coefficient setting routine, FIG. 5 is a time chart showing the relationship between the O2 sensor output voltage and the air-fuel ratio feedback correction coefficient, FIG. 6 is a flowchart of an air-fuel ratio learning value learning routine, and FIG. FIG. 8 is a flowchart of an air pressure ratio learning value table, FIG. 8 is a flowchart of an atmospheric pressure region determination routine, FIG. 9 is a flowchart of a purge execution permission routine, and FIG. 10 is a flowchart of a purge control routine.
[0015]
First, the overall configuration of the engine will be described with reference to FIG. In this figure, reference numeral 1 denotes an engine, and in this embodiment, a horizontally opposed type in which a cylinder block 1a is divided into two banks (left bank on the right side and right bank on the left side in the figure) around a crankshaft 1b. 4 shows a four-cylinder engine. Cylinder heads 2 are provided in both left and right banks of a cylinder block 1a of the engine 1, respectively, and each cylinder head 2 is formed with an intake port 2a and an exhaust port 2b.
[0016]
An intake manifold 3 is communicated with the intake port 2a. A throttle chamber 5 is communicated with the intake manifold 3 through an air chamber 4 in which intake passages of respective cylinders are gathered. An air cleaner 7 is mounted via the air cleaner 7 and communicates with the air intake chamber 8. An exhaust pipe 10 communicates with the exhaust port 2b via an exhaust manifold 9, and a catalyst converter 11 is interposed in the exhaust pipe 10 and communicates with a muffler 12.
[0017]
The throttle chamber 5 is provided with a throttle valve 5a linked to an accelerator pedal, and a bypass passage 13 for bypassing the throttle valve 5a is branched from the intake pipe 6. The bypass passage 13 is provided with an idle speed control valve (ISC valve) 14 that controls the amount of bypass air flowing through the bypass passage 13 during idling to control the idle speed. In the present embodiment, the valve opening of the ISC valve 14 is adjusted according to the duty ratio of a control signal output from an electronic control unit 50 (see FIG. 2) described later.
[0018]
An injector 15 is disposed immediately upstream of the intake port 2 a of each cylinder of the intake manifold 3, and communicates with a fuel tank 17 via a fuel supply path 16. The fuel tank 17 is provided with an in-tank type fuel pump 18, and the fuel from the fuel pump 18 is pressure-fed to the injector 15 and the pressure regulator 20 via a fuel filter 19 interposed in a fuel supply path 16, The fuel is returned from the pressure regulator 20 to the fuel tank 17 and the fuel pressure to the injector 15 is adjusted to a predetermined pressure.
[0019]
Further, from the upper part of the fuel tank 17, a discharge passage 21 for discharging the fuel vapor generated in the fuel tank 17 extends, and a canister 23 having an adsorbing portion made of activated carbon or the like via a two-way valve 22. The upper part is communicated. The canister 23 is provided with a fresh air introduction port communicating with the atmosphere at a lower portion, and a purge passage 24 for introducing a mixture (evaporation gas) of fresh air from the fresh air introduction port and the evaporated fuel gas stored in the adsorption section. It extends from the top.
[0020]
The purge passage 24 is provided with a canister purge control (CPC) duty solenoid valve 25 as an actuator for adjusting the purge amount of the evaporative gas in the middle of the purge passage 24. The purge passage 24 is located downstream of the throttle valve 5 a on the right bank side of the air chamber 4. It is connected to the part and opened. Note that, like the ISC valve 14, the valve opening of the CPC duty solenoid valve 25 is adjusted in accordance with the duty ratio of a control signal output from an electronic control unit 50 (see FIG. 2) described later. On the other hand, an ignition plug 26 that exposes a discharge electrode at the tip end to the combustion chamber 1c is attached to each cylinder of the cylinder head 2, and an ignition coil 27 containing an igniter 28 is connected to the ignition plug 26.
[0021]
Next, sensors for detecting the engine operating state will be described. Immediately downstream of the air cleaner 7 of the intake pipe 6, a thermal intake air amount sensor 29 using a hot wire or a hot film is interposed, and a throttle valve 5 a provided in the throttle chamber 5 is provided with a throttle opening degree. A throttle sensor 30 having a built-in sensor 30a and an idle switch 30b that is turned on when the throttle is fully closed is provided in series. Further, the air chamber 4 is provided with an intake pipe pressure sensor 31 for detecting the intake pipe pressure downstream of the throttle valve 5a as an absolute pressure.
[0022]
In addition, a knock sensor 32 is attached to the cylinder block 1a of the engine 1, a cooling water temperature sensor 34 faces a cooling water passage 33 that communicates the left and right banks of the cylinder block 1a, and further, an empty space is provided upstream of the catalytic converter 11. An O2 sensor 35 as a fuel ratio sensor is provided. A crank angle sensor 37 is provided on the outer periphery of a crank rotor 36 axially mounted on the crankshaft 1b of the engine 1, and further connected to a camshaft 1d that rotates half of the crankshaft 1b. A cylinder discriminating sensor 39 for discriminating the current combustion stroke cylinder, the fuel injection target cylinder, and the ignition target cylinder is provided in a pair.
[0023]
Next, an electronic control unit (ECU) 50 that controls the engine 1 will be described with reference to FIG. The ECU 50 mainly includes a microcomputer in which a CPU 51, a ROM 52, a RAM 53, a backup RAM 54, a counter / timer group 55, and an I / O interface 56 are connected to each other via a bus line, and supplies a stabilized power to each unit. Peripheral circuits such as a constant voltage circuit 57, a drive circuit 58 connected to the I / O interface 56, and an A / D converter 59 are built in.
[0024]
The counter / timer group 55 includes various counters such as a free-run counter, a counter for counting the input of a cylinder discrimination sensor signal (cylinder discrimination pulse), a fuel injection timer, an ignition timer, and a periodic interrupt for generating a periodic interrupt. Timers for counting input intervals of crank angle sensor signals (crank pulses) and watchdog timers for monitoring system abnormalities are collectively referred to for convenience. In addition, various software counter timers are used. Can be
[0025]
The constant voltage circuit 57 is connected to the battery 61 via the first relay contact of the power supply relay 60 having two circuit relay contacts, and is also directly connected to the battery 61, and the ignition switch 62 is turned on. When the contact of the power supply relay 60 is closed, power is supplied to each unit in the ECU 50, and backup power is always supplied to the backup RAM 54 regardless of whether the ignition switch 62 is ON or OFF. Further, the fuel pump 18 is connected to the battery 61 via a relay contact of the fuel pump relay 63. A power supply line for supplying power from the battery 61 to each actuator is connected to the second relay contact of the power supply relay 60.
[0026]
The input port of the I / O interface 56 is connected to an ignition switch 62, an idle switch 30b, a knock sensor 32, a crank angle sensor 37, a cylinder discrimination sensor 39, a vehicle speed sensor 40, and the like. Via the converter 59, the intake air amount sensor 29, the throttle opening sensor 30a, the intake pipe pressure sensor 31, the cooling water temperature sensor 34, the O2 sensor 35, the atmospheric pressure sensor 41, and the like are connected, and the battery voltage VB is reduced. Input and monitored.
[0027]
On the other hand, the output ports of the I / O interface 56 are connected to the relay coils of the power supply relay 60 and the fuel pump relay 63, the ISC valve 14, the injector 15, the CPC duty solenoid valve 25, and the like via the drive circuit 58. And an igniter 28 is connected.
[0028]
The CPU 51 processes detection signals from sensors and switches, which are input via the I / O interface 56, battery voltage, and the like, according to a control program stored in the ROM 52, and various data stored in the RAM 53; Based on various learning value data such as the air-fuel ratio learning value stored in the backup RAM 54, fixed data stored in the ROM 52, etc., the fuel injection amount, ignition timing, control duty for the ISC valve 14, and control duty for the CPC duty solenoid valve 25 The control duty and the like are calculated, and engine control such as air-fuel ratio control (fuel injection control), ignition timing control, idle speed control, and canister purge control are performed.
[0029]
In such an engine control, the ECU 50 accurately grasps the progress of the air-fuel ratio learning based on the number of regions determined to have completed the learning of the air-fuel ratio in the air-fuel ratio control, and determines that the learning has been completed. Only when the determined number of areas reaches the set number of areas, the execution of the canister purge is permitted to achieve both the air-fuel ratio learning control and the canister purge control.
[0030]
The state in which the air-fuel ratio learning is completed is not determined based on the number of updates of the air-fuel ratio learning value, but is determined that learning is progressing in the entire operation region, the change in the air-fuel ratio learning value is small, and stable. In this embodiment, the learning value of the air-fuel ratio in the backup RAM 54 is not cleared, the learning value of the air-fuel ratio does not fluctuate due to a change in the atmospheric pressure, and is set based on the output of the O2 sensor 35. When the average value of the air-fuel ratio feedback correction coefficient is within the set range, it is determined that the learning has been completed.
[0031]
That is, the ECU 50 has the functions of an air-fuel ratio learning unit, a purge control unit, an air-fuel ratio learning completion determination unit, and a purge execution permission unit. Specifically, the functions of each unit are realized by a routine shown in FIG. . Hereinafter, processing related to engine control by the ECU 50 will be described with reference to flowcharts in FIG.
[0032]
FIG. 3 shows a fuel injection control routine that is executed at predetermined intervals after the system is initialized. First, in step S101, a basic fuel injection that determines a basic fuel injection amount corresponding to an intake air amount per stroke. Calculate the pulse width Tp. In this embodiment, a basic fuel injection pulse is obtained by multiplying a value obtained by dividing an intake air amount Q directly measured by the intake air amount sensor 29 by an engine speed NE based on a signal from the crank angle sensor 37 by an injector characteristic constant K. The width Tp is set (Tp ← K × Q / NE).
[0033]
Next, proceeding to step S102, the basic fuel injection pulse width Tp is multiplied or added by various correction terms to calculate a fuel injection pulse width Ti that determines the final fuel injection amount injected from the injector 15. As the correction multiplication term for the basic fuel injection pulse width Tp, for example, various increase correction coefficients COEF relating to cooling water temperature correction, acceleration / deceleration correction, full-open increase correction, post-idle increase correction, etc., and air-fuel ratio correction based on the output of the O2 sensor 35 And an air-fuel ratio learning correction coefficient KBRRC set by interpolation by searching for an air-fuel ratio learning value KLR in an air-fuel ratio learning value table to be described later. There is a voltage correction pulse width Ts for correcting the invalid injection time of the injector 15 that changes according to the voltage. The fuel injection pulse width Ti is calculated by correcting the basic fuel injection pulse width Tp with these correction terms (Ti ← Tp × COEF × LAMBDA × KBLRC + Ts). Then, in step S103, the fuel injection pulse width Ti is set in the injection timer, and the routine exits.
[0034]
Next, the air-fuel ratio feedback correction coefficient setting routine of FIG. 4 for setting the air-fuel ratio feedback correction coefficient LAMBDA will be described.
[0035]
This air-fuel ratio feedback correction coefficient setting routine is executed at predetermined intervals. First, in step S201, it is determined whether an F / B condition (air-fuel ratio feedback condition) is satisfied. The F / B condition is that the output voltage VO2 of the O2 sensor 35 is in an engine non-rotation state of NE = 0, in an engine cranking state, and in an initial state in which any of the conditions after the engine start within a set time is satisfied. It is determined that the F / B condition is not satisfied when the engine is in the inactive state in which the state of the predetermined range does not reach the set value and the state of the predetermined range does not continue for the set time or more, or when the engine is in a transient operation state such as during acceleration / deceleration or during fuel cut. If neither of the conditions is satisfied, it is determined that the F / B condition is satisfied.
[0036]
If the F / B condition is not satisfied, the process proceeds from step S201 to step S202, in which a first-time inversion discrimination flag FR described later is cleared to 0 and initialized (FR ← 0). In step S203, the air-fuel ratio feedback correction is performed. The coefficient LAMBDA is set to LAMBDA = 1.0, and the routine exits. That is, when the F / B condition is not satisfied, the air-fuel ratio feedback correction coefficient LAMBDA is fixed or clamped to 1.0, the air-fuel ratio feedback correction based on the output of the O2 sensor 35 is stopped, and open loop control is performed.
[0037]
On the other hand, when the F / B condition is satisfied in step S201, the process proceeds from step S201 to step S204 and thereafter, where the output voltage VO2 of the O2 sensor 35 and the slice level SL which is a determination value for determining the air-fuel ratio state are determined. According to the comparison result, the air-fuel ratio feedback correction coefficient LAMBDA is set by proportional integral control (PI control).
[0038]
That is, in step S204, the output voltage VO2 of the O2 sensor 35 is read, and the O2 sensor output voltage VO2 is compared with the slice level SL to determine whether the current air-fuel ratio is shifted to the rich side or lean side with respect to the stoichiometric ratio. To determine When VO2 ≧ SL and the air-fuel ratio is shifted to the rich side, the process proceeds to step S205, and the value of the first reversal determination flag FR is referred to.
[0039]
The reversal first determination flag FR is a flag for determining the first time when the air-fuel ratio is reversed from lean to rich or the first time when the air-fuel ratio is reversed from rich to lean, and the air-fuel ratio is changed from lean to rich. 1 → 0 after inversion, and 0 → 1 after inversion from the rich side to the lean side.
[0040]
Therefore, when the air-fuel ratio is rich and FR = 1, it is the first time that the air-fuel ratio has been reversed from lean to rich, so the process proceeds from step S205 to step S206, and as shown in FIG. The correction coefficient LAMBDA is skipped in the negative direction by the proportionality constant PD of the PI control (LAMBDA ← LAMBDA−PD), the inversion initial discrimination flag FR is cleared in step S208 (FR ← 0), and the routine exits.
[0041]
On the other hand, when the air-fuel ratio is on the rich side and FR = 0, skipping in the negative direction by the proportionality constant PD has already been performed on the air-fuel ratio feedback correction coefficient LAMBDA, and the process proceeds from step S205 to step S207. As shown in FIG. 5, the air-fuel ratio feedback correction coefficient LAMBDA is gradually reduced by the integral constant ID of the PI control every time the routine is executed (LAMBDA ← LAMBDA-ID), and the routine exits through the above-described step S308.
[0042]
On the other hand, in step S204, when VO2 <SL and the air-fuel ratio is on the lean side, the process proceeds from step S204 to step S209, and similarly, the value of the first reversal determination flag FR is referred to. When the air-fuel ratio is lean and FR = 0, it is the first time that the air-fuel ratio has been inverted from the rich side to the lean side, so the process proceeds to step S210, where the air-fuel ratio feedback correction coefficient LAMBDA is set as shown in FIG. Skipping is performed in the plus direction by the proportional constant PU (LAMBDA ← LAMBDA + PU), and in step S212, the inversion initial discrimination flag FR is set to 1 (FR ← 1), and the routine exits.
[0043]
When the air-fuel ratio is lean and FR = 1, since the skipping in the plus direction by the proportional constant PU has already been performed on the air-fuel ratio feedback correction coefficient LAMBDA, the process proceeds from step S209 to step S211. As shown in FIG. 5, the air-fuel ratio feedback correction coefficient LAMBDA is gradually increased each time the routine is executed by the integral constant IU of PI control (LAMBDA ← LAMBDA + IU), and the process exits from the routine through the above-described step S212.
[0044]
The deviation of the air-fuel ratio feedback correction coefficient LAMBDA from the reference value is learned by the air-fuel ratio learning value learning routine of FIG. 6, and the learning result is reflected in the above-described fuel injection control routine. In other words, the air-fuel ratio learning correction coefficient KBRRC based on the air-fuel ratio learning value KLR is used to determine the air-fuel ratio due to variations during production of the intake air amount measurement system such as the intake air amount sensor 29 and the fuel supply system such as the injector 15 and deterioration over time. The deviation is absorbed.
[0045]
The air-fuel ratio learning value learning routine is executed at predetermined intervals. First, in step S301, it is determined whether an F / B condition is satisfied. If the F / B condition is satisfied, the process proceeds from step S301 to step S302. If the F / B condition is not satisfied, the process jumps to step S305 to count the number of rich / lean switching operations described later. Clear the value COUNT (COUNT ← 0) and exit the routine.
[0046]
In step S302, it is determined whether or not the current operation region based on the engine speed NE and the basic fuel injection pulse width Tp is within the range of the steady state determination matrix shown in FIG. 7 (NE0 ≦ NE ≦ NEn and Tp0 ≦ Tp ≦ Tpn). Find out what. If it is out of the range of the steady state determination matrix, the process exits the routine through the above-described step S305. If it is within the range of the steady state determination matrix, in step S303, the partition position D1 in the steady state determination matrix is specified. Then, it is checked whether or not the area data NEW of the section position D1 is specified at the time of the previous execution of the routine and is the same as the area data OLD stored in the RAM 53.
[0047]
As a result, when the previous area data OLD and the current area data NEW are different, that is, when the execution of the routine is the first time, or when the operation area at the time of execution of the previous routine and the operation area at the time of execution of the current routine are the same. If it is not in the steady operation state, the process proceeds from step S303 to step S304, the current area data NEW is stored in the RAM 53 as the old data OLD (OLD ← NEW), and the routine exits through the above-described step S305.
[0048]
On the other hand, if the previous area data OLD is the same as the current area data NEW, the process proceeds from step S303 to step S306 to determine whether there is a rich / lean switch of the output voltage VO2 of the O2 sensor 35 within the predetermined time T0. That is, it is determined whether the air-fuel ratio has been inverted from the rich side to the lean side or from the lean side to the rich side. If there is no inversion of the output voltage VO2 of the O2 sensor 35 within the predetermined time T0, the process branches from step S306 to the above-described step S305 to exit the routine, and the output voltage VO2 of the O2 sensor 35 within the predetermined time T0. If there is an inversion, the process proceeds from step S306 to step S307 to count up the count value COUNT (COUNT ← COUNT + 1).
[0049]
Thereafter, the process proceeds to step S308, where it is determined whether or not the count value COUNT has become equal to or greater than the set number of times COUNTSET. If COUNT <COUNTSET, it is determined that the vehicle is not in a steady state, and the routine exits. If COUNT ≧ COUNTSET, that is, the engine rotation If the operation regions based on the number NE and the basic fuel injection pulse width Tp are substantially the same, and the number of reversals of the output voltage VO2 of the O2 sensor 35 is equal to or more than COUNTSET at this time, it is determined to be in a steady state, and the process proceeds to step S309.
[0050]
In step 309, the count value COUNT is cleared (COUNT ← 0). Then, in step S310, the air-fuel ratio feedback correction is performed while the output voltage VO2 of the O2 sensor 35 crosses the slice level for a set number of times (for example, three times). An average value LAMBDAAVE of the maximum value and the minimum value of the coefficient LAMBDA is calculated, and a deviation amount (deviation from the reference value) ΔLAMBDA of the average value LAMBDAAVE from the reference value 1.0 is calculated (ΔLAMBDA ← LAMBDAAVE−1.0). .
[0051]
Then, the process proceeds to step S311 to search for the air-fuel ratio learning value KLR from the air-fuel ratio learning value table of the backup RAM 54 using the engine speed NE and the basic fuel injection pulse width Tp as parameters. As shown in FIG. 7, the air-fuel ratio learning value table includes an air-fuel ratio feedback correction coefficient LAMBDA in a steady operation state for each region specified by the engine speed NE and the basic fuel injection pulse width Tp as the engine load. The air-fuel ratio learning value KLR is stored based on the difference between the average value and the reference value during a predetermined number of air-fuel ratio rich / lean iterations. In an unlearned initial state, all the data in the table are initial values ( = 1.0).
[0052]
Accordingly, when the process proceeds from step S311 to step S312, an air-fuel ratio learning value KLR is set from the searched air-fuel ratio learning value KLR and the deviation amount ΔLAMBDA calculated in step S310 (KLR ← KLR + M × ΔLAMBDA; The air-fuel ratio learning value KLR is written to the corresponding address of the air-fuel ratio learning value table, and the routine exits.
[0053]
In parallel with the fuel injection control routine, the air-fuel ratio feedback correction coefficient setting routine, and the air-fuel ratio learning value learning routine, the atmospheric pressure region determination routine in FIG. 8, the purge execution permission routine in FIG. 9, and the purge control routine in FIG. Are executed at predetermined intervals. In the atmospheric pressure region determination routine, it is determined to which of a plurality of previously divided atmospheric pressure regions the current atmospheric pressure belongs. Based on the determination result of the atmospheric pressure and the change of the air-fuel ratio feedback correction coefficient LAMBDA, the purge is performed. The execution permission routine determines whether the execution of the canister purge is permitted. When the execution of the canister purge is permitted, the purge control routine executes the canister purge.
[0054]
First, the atmospheric pressure region determination routine of FIG. 8 will be described. In the present embodiment, the atmospheric pressure region is divided into three regions 1, 2, and 3 with hysteresis as described below, and it is determined to which region the current atmospheric pressure PAT belongs.
(1) Atmospheric pressure area 1 ... PAT ≧ P1
(2) Atmospheric pressure region 2 ... P2 ≦ PAT <P1
(3) Atmospheric pressure region 3 ... PAT <P2
[0055]
Therefore, in the atmospheric pressure region determination routine, in step S401, it is determined whether the current atmospheric pressure PAT detected by the atmospheric pressure sensor 41 is equal to or higher than P1 (for example, 690 mmHg = 91.99 kPa). If PAT ≧ P1, the routine proceeds to step S402 with the area number of the atmospheric pressure area set to 1, and if PAT <P1, the processing proceeds from step S401 to step S403, where the current atmospheric pressure PAT is P2 ( For example, it is checked whether it is less than 620 mmHg = 82.66 kPa).
[0056]
As a result, if PAT ≧ P2, the process proceeds from step S403 to step S404, where the area number of the atmospheric pressure area is set to 2, and the routine exits. On the other hand, if PAT <P2, the process proceeds from step S403 to step S405, where the area number of the atmospheric pressure area is set to 3 and the routine exits.
[0057]
In the purge execution permission routine of FIG. 9, in the first step S501, it is determined whether or not the contents of the backup RAM 54 have been cleared, based on whether or not the keyword has been cleared. This keyword is specific data (for example, 2-bit “11” data) written at a specific address in the backup RAM 54. When the specific data is cleared, the battery 61 is repaired or the like. Is removed, the backup power supply to the backup RAM 54 is cut off, and the learning value data in the backup RAM 54 is destroyed, or the learning value data in the backup RAM 54 is deleted by an intentional clearing operation of a keyword for diagnosis or the like. Judge that it has been initialized.
[0058]
If the keyword in the backup RAM 54 has been cleared, the process proceeds from step S501 to step S502 to clear a learning completion determination described later. Then, in step S503, the purge execution permission flag FLG for instructing the execution of the canister purge is cleared (FLG ← 0), the execution of the canister purge is prohibited, and the routine exits.
[0059]
It should be noted that the keyword in the backup RAM 54 can be reset at the dealer's service station or the like after being cleared, but in addition, for example, the air-fuel ratio learning value KLR can be set by the above-described air-fuel ratio learning value learning routine. When the battery 61 has been updated a predetermined number of times, the battery 61 is automatically reset to prevent the canister purge from being disabled even when the user removes and replaces the battery 61 by himself / herself.
[0060]
On the other hand, if the keyword in the backup RAM 54 has not been cleared, the process advances from step S501 to step S504 to check whether the atmospheric pressure region has changed. As a result, in step S504, when the atmospheric pressure region number at the time of execution of the previous routine and the atmospheric pressure region number at the time of execution of the current routine are different, and the atmospheric pressure region is changing, If the driving environment changes, the air-fuel ratio learning value is expected to fluctuate due to a change in the volumetric efficiency of the engine due to a change in the intake air density and the exhaust pressure. , S503, and exits the routine.
[0061]
If the atmospheric pressure area number at the time of executing the previous routine and the atmospheric pressure area number at the time of executing the current routine are the same and the atmospheric pressure area has not changed, the process proceeds from step S504 to step S505, and the current learning area It is determined whether or not the average value LAMBDAAVE of the air-fuel ratio feedback correction coefficient LAMBDA in the above is within a set range (for example, −5% ≦ LAMBDAAVE ≦ 5%) with respect to the reference value (1.0).
[0062]
If the average value LAMBDAAVE of the air-fuel ratio feedback correction coefficient LAMBDA is out of the set range, it is determined that the learning in the current learning region has not been completed, and the process exits the routine via the above-described step S503. On the other hand, when the average value LAMBDAAVE of the air-fuel ratio feedback correction coefficient LAMBDA is within the set range, the process proceeds from step S505 to step S506, and it is determined that the learning of the current learning region is completed.
[0063]
That is, in the state where the air-fuel ratio learning is not completed, the air-fuel ratio correction is performed using the air-fuel ratio feedback correction coefficient LAMBDA based on the output of the O2 sensor 35, so the air-fuel ratio feedback correction coefficient LAMBDA is corrected with a large value. In some cases, the air-fuel ratio learning is incomplete and in progress, and when the fluctuation of the air-fuel ratio feedback correction coefficient LAMBDA becomes small, it can be determined that the air-fuel ratio learning has been completed.
[0064]
Therefore, in order to avoid the influence of the transient air-fuel ratio fluctuation, the learning completion determination is performed using the average value LAMBDAAVE of the air-fuel ratio feedback correction coefficient LAMBDA. When the average value LAMBDAAVE is within the set range, the learning of the corresponding area is completed. It is determined that it has been done. An appropriate range of the average value LAMBDAAVE of the air-fuel ratio feedback correction coefficient LAMBDAAVE is determined in advance for each learning region by simulation or experiment in consideration of the engine characteristics, and is stored in the ROM 52 as fixed data.
[0065]
Thereafter, the process proceeds from step S506 to step S507, and it is determined whether or not the number of regions determined to be learning completed (the number of learning completion determination regions) is equal to or larger than the set region number. This set number of regions is a state in which the progress of the air-fuel ratio learning can sufficiently reflect the air-fuel ratio learning value to the fuel injection amount even if the progress of the air-fuel ratio learning executes the canister purge without decreasing the execution frequency of the canister purge. , For example, three regions.
[0066]
If the number of learning completion determination areas is less than the set number, the routine exits the routine while maintaining the purge execution permission flag FLG in the clear state (purging execution prohibited) in step S503 described above. If so, the process proceeds from step S507 to step S508 to set a purge execution permission flag FLG to permit execution of the canister purge (FLG ← 1), and exit the routine.
[0067]
This purge execution permission flag FLG is referred to in the purge control routine of FIG. 10. When FLG = 1, the CPC duty solenoid valve 25 is driven to open, and the evaporated fuel is purged from the canister 23 to the intake system.
[0068]
That is, in the purge control routine, the value of the purge execution permission flag FLG is referred to in the first step S601 to check whether or not the execution of the canister purge is permitted. If the execution of the canister purge is prohibited with FLG = 0, the control duty (duty ratio) CPCD for the CPC duty solenoid valve 25 is set to 0 in step S602, and the routine exits. Thus, when the canister purge of FLG = 0 is prohibited, the CPC duty solenoid valve 25 is fully closed by the control duty CPCD = 0, and the canister purge is prohibited. On the other hand, if FLG = 0 and execution of the canister purge is permitted, the process proceeds to step S603.
[0069]
In step S603, a basic control duty CPCDMAP is set with reference to the control duty map MAP based on the intake pipe pressure PB and the intake air amount Q (CPDMAMAP ← MAP (PB, Q)). As exemplified in step S603, the control duty map MAP obtains an optimum purge amount by using the intake pipe pressure PB and the intake air amount Q as parameters in a state where the above-described purge conditions are satisfied by simulation or experiment in advance. The control duty (duty ratio) for the CPC duty solenoid valve 25 is obtained, and this is stored in the ROM 52 as the basic control duty CPCMDAP.
[0070]
After setting the basic control duty CPCDMAP in step S603, the process proceeds to step S604, in which the basic control duty CPCDAMP is corrected by various purge correction coefficients CPCK corresponding to the operation state, and the control duty (duty ratio) CPCD for the CPC duty solenoid valve 25 is set. Is set (CPCD ← CPDMAP × CPCK). Then, in step S605, the control duty CPCD is set as the duty ratio of the drive signal for the CPC duty solenoid valve 25, and the routine exits. Therefore, when FLG = 1 and the canister purge is permitted, the control duty CPCD is set according to the engine operating state at that time, and the purge is executed by opening the CPC duty solenoid valve 25.
[0071]
As described above, in the present embodiment, for each learning region, a change in the traveling environment due to the atmospheric pressure region and a change in the average value of the air-fuel ratio feedback correction coefficient LAMBDA are checked to determine the learning completion, and it is determined that the learning is completed. The number of regions reaches a state in which the progress of the air-fuel ratio learning can sufficiently reflect the air-fuel ratio learning value to the fuel injection amount even if the progress of the air-fuel ratio learning is performed without reducing the execution frequency of the canister purge. When the number of regions determined to have been reached has reached, the execution of the canister purge is permitted, so that the air-fuel ratio learned value can be reflected in the fuel injection amount at the time of the air-fuel ratio correction accompanying the execution of the canister purge. It is possible to improve the air-fuel ratio controllability by absorbing the variation in the tolerance, thereby contributing to the improvement of the exhaust gas emission and the reduction of the fuel consumption.
[0072]
【The invention's effect】
As described above, according to the present invention, it is possible to reliably determine whether or not the air-fuel ratio learning has sufficiently proceeded, to improve the controllability by making the air-fuel ratio learning control and the canister purge control compatible. .
[Brief description of the drawings]
FIG. 1 is an overall view of an engine system.
FIG. 2 is a circuit configuration diagram of an electronic control system.
FIG. 3 is a flowchart of a fuel injection control routine.
FIG. 4 is a flowchart of an air-fuel ratio feedback correction coefficient setting routine.
FIG. 5 is a time chart showing a relationship between an O2 sensor output voltage and an air-fuel ratio feedback correction coefficient.
FIG. 6 is a flowchart of an air-fuel ratio learning value learning routine.
FIG. 7 is an explanatory diagram of a steady state determination matrix and an air-fuel ratio learning value table.
FIG. 8 is a flowchart of an atmospheric pressure region determination routine.
FIG. 9 is a flowchart of a purge execution permission routine.
FIG. 10 is a flowchart of a purge control routine.
[Explanation of symbols]
1 engine
17 Fuel tank
23 Canister
50 electronic control unit (air-fuel ratio learning means, purge control means, air-fuel ratio learning completion determination means, purge execution permission means)
KLR air-fuel ratio learning value
LAMBDA air-fuel ratio feedback correction coefficient
LAMBDAAVE Average value of air-fuel ratio feedback correction coefficient
PAT atmospheric pressure

Claims (2)

Air-fuel ratio learning means for learning a deviation from a reference value of an air-fuel ratio feedback correction coefficient set based on an output of an air-fuel ratio sensor interposed in an exhaust system of an engine, and a canister for storing evaporative fuel in a fuel tank. A purge control means for controlling an amount of fuel vapor to be purged to an intake system of the engine.
Air-fuel ratio learning completion determining means for determining whether or not air-fuel ratio learning has been completed for each learning region by the air-fuel ratio learning means;
Purge execution permission means for permitting the purge control means to perform purge of evaporated fuel from the canister to the intake system when the number of regions determined to be completed by the air-fuel ratio learning completion determination means is equal to or greater than a set number. A control device for an engine. The air-fuel ratio learning completion determination means includes:
In the current learning region, when the average value of the air-fuel ratio feedback correction coefficient is within the set range without changing the atmospheric pressure between a plurality of pre-divided atmospheric pressure regions, it is determined that the air-fuel ratio learning is completed. The engine control device according to claim 1, wherein:

Images