JP2004019467A - Engine air-fuel ratio feedback control device - Google Patents

Abstract

An air-fuel ratio feedback correction is performed in accordance with a degree of deterioration of a catalyst to prevent deterioration of exhaust emission even when the deterioration of the catalyst progresses. The catalyst diagnosis learning value bCATDIAGA is updated with the diagnosis value CATDIGA (S17), and based on the catalyst diagnosis learning value bCATDIAGA, the I component rich correction value bIRCATSBFBA is set with reference to the I component rich correction value setting table tIRCATSBFBA (S18). Further, the I-minute lean correction value bILCATSBFBA is set with reference to the I-minute lean correction value setting table tILCATSBFBA (S19). Then, the I component rich constant and the I component lean constant of the air-fuel ratio feedback correction coefficient are corrected using the two correction values bIRCATSBFBA and bILCATSBFBA.
[Selection diagram] FIG.

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an air-fuel ratio feedback control device for an engine, and more particularly to an air-fuel ratio feedback control device for an engine that corrects an air-fuel ratio feedback correction coefficient according to the degree of deterioration of a catalyst.
[0002]
[Prior art]
Generally, in the air-fuel ratio control of the engine, the air-fuel ratio state is detected by the output of an air-fuel ratio sensor (O2 sensor) disposed upstream of the catalyst, and the air-fuel ratio is feedback-controlled. In order to improve the deterioration of the air-fuel ratio control accuracy due to the variation of the value characteristic and the deterioration of the fuel injection valve and the like with the passage of time, O2 sensors are disposed upstream and downstream of the catalyst, respectively. Various so-called double O2 sensor (DOS) systems that control the air-fuel ratio based on the value have been proposed.
[0003]
In the air-fuel ratio control by the DOS system, for example, as disclosed in Japanese Patent Application Laid-Open No. 62-29738, the output of a second air-fuel ratio sensor (rear O2 sensor) disposed downstream of the catalyst is controlled. -Fuel ratio feedback control parameter calculated in accordance with the air-fuel ratio feedback control parameter R calculated based on the output of the first air-fuel ratio sensor (front O2 sensor or wide area air-fuel ratio (A / F) sensor) disposed upstream of the catalyst An air-fuel ratio feedback correction coefficient is calculated according to F, and the fuel injection amount is corrected by the air-fuel ratio using the air-fuel ratio feedback correction coefficient.
[0004]
The air-fuel ratio feedback control parameter R is set by proportional integral control (PI control) by comparing the output value of the rear O2 sensor with the rich / lean inversion level. The proportional constant (P constant) is set to a unique value.
[0005]
[Problems to be solved by the invention]
Here, when the catalyst is new, the output value FVO2 [V] of the front O2 sensor shown in FIG. 18 (a) changes from rich (the output value FVO2 [V] is larger than the rich / lean determination line) to lean (output). The value FVO2 [V] is smaller than the rich / lean determination line), and the inversion cycle of the output value RVO2 [V] of the rear O2 sensor is the same as the inversion cycle from lean to rich due to the O2 storage effect of the catalyst. As shown in FIG. Then, as the catalyst deteriorates, the O2 storage effect of the catalyst decreases, and the inversion cycle of the output value RVO2 [V] of the rear O2 sensor decreases as shown in FIG.
[0006]
Therefore, when the I-part constant and the P-part constant that determine the air-fuel ratio control amount per unit time are set in accordance with the case where the catalyst is new and the inversion cycle of the output value RVO2 [V] of the rear O2 sensor is long, When the reversal cycle of the output value RVO2 [V] of the rear O2 sensor is shortened due to the deterioration of the catalyst, the air-fuel ratio is not sufficiently corrected, and the exhaust emission is disadvantageously deteriorated.
[0007]
Further, if the I component constant and the P component constant are set in accordance with the case where the deterioration of the catalyst progresses and the reversal cycle of the output value RVO2 [V] of the rear O2 sensor is shortened, the output of the rear O2 sensor is new when the catalyst is new. When the reversal cycle of the value RVO2 [V] is long, an excessive air-fuel ratio correction is performed in a short time, and there is a disadvantage that the exhaust emission deteriorates.
[0008]
In view of the above circumstances, the present invention performs the air-fuel ratio feedback correction according to the degree of deterioration of the catalyst, so that not only when the catalyst is new, but also when the deterioration of the catalyst progresses, the exhaust emission deteriorates. It is an object of the present invention to provide an air-fuel ratio feedback control device for an engine, which can prevent the occurrence of an air-fuel ratio and efficiently utilize the O2 storage effect of a catalyst.
[0009]
[Means for Solving the Problems]
According to a first aspect of the present invention, a front air-fuel ratio sensor and a rear air-fuel ratio sensor are disposed upstream and downstream of a catalyst interposed in an exhaust system of an engine, respectively. An air-fuel ratio feedback correction coefficient is set by proportional integral control according to the air-fuel ratio feedback correction coefficient, and the air-fuel ratio feedback correction coefficient is used to correct the fuel injection amount with the air-fuel ratio feedback correction coefficient. Reversal cycle calculating means for detecting a lean reversal cycle; catalyst deterioration diagnostic value setting means for setting a catalyst deterioration diagnosis value based on the reversal cycle; and a proportional constant and integral of an air-fuel ratio feedback correction coefficient based on the catalyst deterioration diagnosis value Correction means for setting a correction value for correcting at least one of the constants.
[0010]
In such a configuration, a rich / lean reversal cycle of the output value of the rear air-fuel ratio sensor is detected, a catalyst deterioration diagnosis value is set based on the reversal cycle, and then an air-fuel ratio feedback correction coefficient is set based on the catalyst deterioration diagnosis value. A correction value for correcting at least one of the proportional constant and the integral constant is set.
[0011]
According to a second aspect of the present invention, a front air-fuel ratio sensor and a rear air-fuel ratio sensor are disposed upstream and downstream of a catalyst interposed in an exhaust system of an engine, respectively, and proportionally provided in accordance with output values of the two air-fuel ratio sensors. In an air-fuel ratio feedback control device for an engine, in which an air-fuel ratio feedback correction coefficient is set by integral control and the fuel injection amount is corrected by the air-fuel ratio feedback correction coefficient, the front air-fuel ratio sensor calculates the amount of change in the output value of the front air-fuel ratio sensor. Air-fuel ratio sensor output value change amount calculation means, rear air-fuel ratio sensor output value change amount calculation means for calculating the change amount of the rear air-fuel ratio sensor output value, change amount of the front air-fuel ratio sensor output value and the rear air Comparison value calculation means for calculating a comparison value with the amount of change in the output value of the fuel ratio sensor, and a catalyst deterioration diagnosis value setting means for setting a catalyst deterioration diagnosis value based on the comparison value When, characterized in that it comprises a correcting means for setting a correction value for correcting at least one of the proportional constant and integral constant of the air-fuel ratio feedback correction coefficient on the basis of the catalyst deterioration diagnostic value.
[0012]
In such a configuration, the amount of change in the output value of the front air-fuel ratio sensor is calculated, and the amount of change in the output value of the rear air-fuel ratio sensor is calculated. A comparison value with the amount of change in the value is calculated, a catalyst deterioration diagnosis value is set based on the comparison value, and at least one of the proportional constant and the integration constant of the air-fuel ratio feedback correction coefficient is corrected based on the catalyst deterioration diagnosis value. Set the correction value.
[0013]
In this case, preferably, 1) the correction means sets an integral rich correction value and an integral lean correction value based on the catalyst deterioration diagnosis value, and corrects the air-fuel ratio feedback correction coefficient with both correction values. Features.
[0014]
2) The correction means sets a proportional rich correction gain and a proportional lean correction gain based on the catalyst deterioration diagnosis value, and sets a proportional rich correction value and a proportional lean correction value based on both correction gains. The air-fuel ratio feedback correction coefficient is corrected by the two correction values.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. 1 to 8 show a first embodiment of the present invention. FIG. 1 shows a schematic configuration diagram of an engine control system.
[0016]
Reference numeral 1 in the figure denotes an engine, which in this embodiment is a horizontally opposed engine. An intake manifold 2 and an exhaust manifold 3 are connected to the engine 1, respectively.
[0017]
The upstream side of the intake manifold 2 is gathered in an air chamber 4, and an intake pipe 5 is communicated upstream of the air chamber 4, and the intake pipe 5 is communicated with an air cleaner (not shown). Further, a throttle valve 6 is interposed immediately upstream of the air chamber 4 of the intake pipe 5. Although not shown, an intake air amount sensor for detecting an intake air amount is provided immediately downstream of the air cleaner in the intake pipe 5. Further, an injector 7 is disposed immediately upstream of the intake port of each cylinder of the intake manifold 2.
[0018]
Further, an exhaust pipe 8 is communicated with the gathering portion of the exhaust manifold 3, and a muffler 9 is communicated downstream of the exhaust pipe 8. Further, a catalyst 10 composed of a three-way catalyst or the like is disposed in the exhaust pipe 8 from the upstream side, and as a front air-fuel ratio sensor that detects rich / lean air-fuel ratio from the oxygen concentration in the exhaust gas upstream of the catalyst 10. A front O2 sensor 12 is provided, and a rear O2 sensor 13 as a rear air-fuel ratio sensor that detects rich / lean air-fuel ratio from the oxygen concentration in the exhaust gas is provided downstream of the catalyst 10.
[0019]
A crank angle sensor 14 for detecting an engine speed from the rotation of the crank rotor is provided on the outer periphery of a crank rotor (not shown) which is mounted on the crank shaft 1a of the engine 1.
[0020]
On the other hand, reference numeral 21 denotes an electronic control unit (ECU), which includes a CPU 22, a ROM (not shown), a RAM (not shown), and a rewritable nonvolatile memory (for example, an EEP (Electrically Erasable Programmable) ROM, It mainly includes a microcomputer having a flash memory) and a backup RAM. The CPU 22 processes detection signals and the like from each sensor according to a control program stored in the ROM, and stores various data stored in the RAM and various learning value data stored in the nonvolatile memory and the backup RAM. , The fuel injection amount, the ignition timing, and the like are calculated, and engine control such as fuel injection control and ignition timing control is performed.
[0021]
In such an engine control system, in the present embodiment, the air-fuel ratio feedback correction coefficient LAMBDA by the DOS system is set by proportional integral (PI) control, and the I component rich constant IR and the I component lean constant that provide the PI control The I component rich correction value bIRCATSBFBA and the I component lean correction value bILCATSBFBA for correcting IL are set according to the rich / lean inversion cycle Ts of the rear O2 sensor 13.
[0022]
That is, the rich / lean reversal cycle Ts of the rear O2 sensor 13 has a correlation with the degree of progress of the deterioration of the catalyst. Is long, the number of reversals in a predetermined time is small. In this case, by performing normal PI control without setting the I-component rich correction value bIRCATSBFBA and the I-component lean correction value bILCATSBFBA, appropriate air-fuel ratio control can be performed. Done.
[0023]
On the other hand, when the oxygen (O2) storage effect is reduced due to the progress of the catalyst deterioration and the inversion cycle Ts of the rear O2 sensor 13 is gradually shortened, the I value rich correction value bIRCATSBFBA and the I value The lean correction value bILCATSBFBA is set, whereby the air-fuel ratio fluctuation due to the air-fuel ratio feedback control increases, and the increase in the air-fuel ratio fluctuation promotes the O2 adsorption and desorption actions of the catalyst, and the exhaust gas accompanying the progress of catalyst deterioration Emission deterioration is prevented.
[0024]
The air-fuel ratio control executed by the ECU 21 is specifically processed according to the flowcharts shown in FIGS.
[0025]
The catalyst deterioration diagnosis routine shown in FIG. 2 is performed only once or every predetermined time, for example, after the engine is started, after the catalyst 10 is activated, and when the engine operating state is in a steady state such as idling. Is executed.
[0026]
First, in step S1, the output value RVO2 [V] of the rear O2 sensor 13 is read. In step S2, the output value RVO2 [V] is compared with the rich / lean determination line (see FIG. 17). It is checked whether the output value RVO2 is on the lean side.
[0027]
If the output value RVO2 of the rear O2 sensor 13 is on the lean side, the process proceeds to step S3, and it is determined whether or not the previous output value RVO2 of the rear O2 sensor 13 is rich. It is determined that the output value RVO2 has been inverted from lean to rich, the process proceeds to step S4, the timer T is cleared (T ← 0), and the routine exits.
[0028]
If the previous output value RVO2 is lean in step S3, the lean state of the air-fuel ratio continues, so the process proceeds to step S5, the timer T is counted up (T ← T + 1), and the routine exits.
[0029]
On the other hand, if it is determined in step S2 that the output value RVO2 of the rear O2 sensor 13 is on the rich side, the process proceeds to step S6, where it is determined whether or not the previous output value RVO2 of the rear O2 sensor 13 is lean. , Since the rich state is continuing, the routine exits as it is. Also, in the case of lean, it is determined that the output value RVO2 has been inverted from rich to lean, the process proceeds to step S7, the inversion cycle Ts is set by the count value of the timer T (Ts ← T), and the process proceeds to step S8 to perform inversion. Based on the cycle Ts, the table is referred to with interpolation calculation, or a catalyst deterioration diagnosis value CATDIGA is set by calculation. The catalyst deterioration diagnosis value CATDIAGA is a value indicating the degree of deterioration of the catalyst 10, and CATDIAGA is set to 0 at the time of a new layer, and as the count value of the timer T decreases, in other words, the deterioration of the catalyst 10 Is set to a larger value as progresses.
[0030]
This catalyst deterioration diagnosis value CATDIGA is read in a catalyst deterioration correction value setting routine shown in FIG. This routine is started every predetermined calculation cycle (32 msec in the present embodiment). First, in step S11, the value of the initialized flag FLGINI stored in the nonvolatile memory or the backup RAM (not shown) is changed. refer. The initialized flag FLGINI is reset (FLAGINI ← 0), for example, when data stored in the nonvolatile memory or the backup RAM is not set, or is destroyed or lost.
[0031]
When FLGINI = 0, it is determined that each set value data stored in the non-volatile memory or the backup RAM (not shown) is unset, destroyed, or lost, and the process proceeds to step S12 and proceeds to step S12. In steps S14 to S14, each set value stored in the nonvolatile memory or the backup RAM (not shown) is initialized.
[0032]
First, in step S12, the catalyst diagnosis learning value bCATDIAGA is initialized by a preset catalyst diagnosis initial value constant kCATDIGA (bCATDIAGA ← kCATDIAGA).
[0033]
In step S13, the I component rich correction value bIRCATSBFBA is initialized with a preset I component rich correction initial value constant kIRCATSBFBA (bIRCATSBFBA ← kIRCATSBFBA).
[0034]
In step S14, the I minute lean correction value bILCATSBFBA is initialized with a preset I minute lean correction initial value constant kILCATSBFBA (bILCATSBFBA ← kILCATSBFBA).
[0035]
Note that the above-described initial value constants kCATDIAGA, kIRCATSBFBA, and kILCATSBFBA are set to values corresponding to, for example, the catalyst 10 at the time of a new product.
[0036]
Then, the process proceeds to step S15, where the initialization flag FLGINI is set (FLGINI ← 1), indicating that the initialization has been completed, and then the routine exits.
[0037]
On the other hand, when it is determined in step S11 that the initialization of FLGINI = 1 has been set, the process proceeds to step S16, where the catalyst deterioration diagnosis value CATDIAGA is read, and the catalyst deterioration diagnosis value CATDIAGA is compared with the catalyst diagnosis learning value bCATDIAGA. .
[0038]
Then, if CATDIAGA ≦ bCATDIAGA, it is determined that the deterioration of the catalyst 10 has not progressed, and the routine exits from the routine. On the other hand, when CATDIAGA> bCATDIAGA, it is determined that the deterioration of the catalyst 10 has progressed, and the process proceeds to step S17 to update the catalyst diagnosis learning value bCATDIAGA with the current catalyst deterioration diagnosis value CATDIAGA (bCATDIAGA ← CATDIAGA).
[0039]
Then, the process proceeds to step S18, and based on the catalyst diagnosis learning value bCATDIAGA updated in step S17, the I component rich correction value setting table tIRCATSBFBA shown in FIG. 6A is referenced with interpolation calculation, and the I component rich correction value bIRCATSBFBA is obtained. Set.
[0040]
In step S19, based on the catalyst diagnosis learning value bCATDIAGA updated in step S17, the I-minute lean correction value setting table tILCATSBFBA shown in FIG. 6B is referenced with interpolation calculation, and the I-minute lean correction value bILCATSBFBA is calculated. Set and exit the routine.
[0041]
As shown in FIGS. 6A and 6B, the I-part rich correction value setting table tIRCATSBFBA and the I-part lean correction value setting table tILCATSBFBA have the catalyst diagnosis learning value bCATDIAGA in relation to the catalyst diagnosis learning value bCATDIAGA. As the value increases, the I-rich correction value bIRCATSBFBA and the I-lean correction value bILCATSBFBA, which are larger, are stored.
[0042]
In the present embodiment, the correction values bIRCATSBFBA and bILCATSBFBA stored in the I-rich correction value setting table tIRCATSBFBA and the I-lean correction value setting table tILCATSBFBA show the same characteristics, but have different characteristics. There may be.
[0043]
Also, the initialized flag FLGINI, the I-rich correction value bIRCATSBFBA, and the I-lean correction value bILCATSBFBA are stored at predetermined addresses in a nonvolatile memory or a backup RAM, and the data is stored even after the ignition switch is turned off. When the ignition switch is turned on next time, the data stored when the ignition switch was turned off last time can be read.
[0044]
The I component rich correction value bIRCATSBFBA and the I component lean correction value bILCATSBFBA are read in the air-fuel ratio feedback correction coefficient setting routine shown in FIG.
[0045]
This routine is executed every set calculation cycle (for example, 32 msec). First, in step S21, it is checked whether the air-fuel ratio feedback control condition is satisfied. As the air-fuel ratio feedback control condition, for example, when the catalyst 10 is in an active state and the engine is in a steady operation state, it is determined that the air-fuel ratio feedback control condition is satisfied.
[0046]
If the air-fuel ratio feedback control condition is not satisfied, the process proceeds to step S22, and the air-fuel ratio feedback correction coefficient LAMBDA is set to 1.0 (LAMBDA ← 1.0). As a result, the air-fuel ratio control when the air-fuel ratio feedback control condition is not satisfied is open-loop control.
[0047]
Thereafter, the process proceeds to step S23, where the feedback control condition satisfaction flag FLGCLF is cleared (FLGCLF ← 0), and the routine exits.
[0048]
On the other hand, when it is determined in step S21 that the air-fuel ratio feedback control condition is satisfied, and the process proceeds to step S24, the air-fuel ratio feedback correction coefficient is obtained by PI control in accordance with the output value FVO2 [V] of the front O2 sensor 12 through the processing in step S24 and subsequent steps. Set LAMBDA.
[0049]
In step S24, the output value FVO2 of the front O2 sensor 12 is compared with the rich / lean reversal level FSL for judging the air-fuel ratio state. When FVO2 ≧ FSL, the process proceeds to step S25, and the process proceeds to step S25. The air-fuel ratio state at the time of execution of the routine is checked. If the previous air-fuel ratio is lean, this is the first execution of the routine after the air-fuel ratio has been inverted from lean to rich, so the process proceeds to step S26, and the next (1) The current air-fuel ratio feedback correction coefficient LAMBDA is calculated from the equation.
LAMBDA ← LAMBDA-PL (1)
Here, PL is a lean component for P and is a fixed value set in advance.
[0050]
As a result, as shown in FIG. 7 (b), the air-fuel ratio feedback correction coefficient LAMBDA is skipped to the lean side by the P-part lean constant PL.
[0051]
On the other hand, when it is determined in step S25 that the previous air-fuel ratio is rich, the process proceeds to step S27 because the air-fuel ratio rich state has been continued, and the I-minute lean correction value bILCATSBFBA is read in step S28. The current air-fuel ratio feedback correction coefficient LAMBDA is calculated from the following equations (2) and (3).
LAMBDA ← LAMBDA-IL (2)
IL ← IO + bILCATSBFBA (3)
Here, IL is an I-minute lean constant. IO is a basic integration constant, which is a fixed value set in advance.
[0052]
By the way, since the initial value of the I minute lean correction value bILCATSBFBA is the I minute lean correction initial value constant kILCATSBFBA, the equation (3) when the catalyst is new is:
IL ← IO + kILCATSBFBA (3 ′)
Can be rewritten as
[0053]
As a result, as shown by the solid line in FIG. 8D, as the deterioration of the catalyst 10 progresses, the air-fuel ratio feedback correction coefficient LAMBDA is gradually reduced by the I-minute lean correction value bILCATSBFBA in each calculation cycle.
[0054]
Thereafter, the process proceeds from step S26 or S28 to step S29, sets the feedback control condition satisfaction flag FLGCLF (FLGCLF ← 1), and exits the routine.
[0055]
On the other hand, if it is determined in step S24 that FVO2 <FSL and the air-fuel ratio is lean, the flow branches to step S31 to check the state of the air-fuel ratio during the previous execution of the routine. Since this is the first routine execution after the fuel ratio is inverted from lean to rich, the process proceeds to step S32, calculates the current air-fuel ratio feedback correction coefficient LAMBDA from the following equation (4), and returns to step S29.
LAMBDA ← LAMBDA + PR (4)
Here, PL is a lean component for P and is a fixed value set in advance.
[0056]
As a result, as shown in FIG. 7B, the air-fuel ratio feedback correction coefficient LAMBDA is skipped toward the rich side by the amount corresponding to the P component rich constant PR.
[0057]
On the other hand, when it is determined in step S31 that the previous air-fuel ratio is lean, the process proceeds to step S33, where the I-component rich correction value bIRCATSBFBA is read, and in step S34, the following formulas (6) and (7) are used. Then, the current air-fuel ratio feedback correction coefficient LAMBDA is calculated, and the process returns to step S29.
LAMBDA ← LAMBDA + IR (6)
IR ← IO + bIRCATSBFBA (7)
Here, IR is an I component rich constant.
[0058]
Note that the initial value of the I component rich correction value bIRCATSBFBA is the I component rich correction initial value constant kIRCATSBFBA.
IR ← IO + kIRCATSBFBA ... (7 ')
Can be rewritten as
[0059]
As a result, as shown by the solid line in FIG. 8 (c), as the deterioration of the catalyst 10 progresses, the air-fuel ratio feedback correction coefficient LAMBDA is gradually increased by the I-minute rich correction value bIRCATSBFBA every calculation cycle.
[0060]
Therefore, as shown in FIG. 7, the air-fuel ratio feedback correction coefficient LAMBDA is set by PI control according to the result of comparison between the output value FVO2 of the front O2 sensor 12 and the rich / lean inversion level FSL.
[0061]
Then, an I-minute lean constant IL at the time of the I-minute lean control in the PI control is set by adding the I-minute lean correction value bILCATSBFBA to the basic integration constant IO, and the air-fuel ratio is determined by the I-minute lean constant IL (= IO + bILCATSBFBA). The feedback correction coefficient LAMBDA is gradually reduced. Further, the I-component rich constant IR at the time of the I-component rich control is set by adding the I-component rich correction value bIRCATSBFBA to the basic integration constant IO, and the air-fuel ratio feedback correction coefficient LAMBDA is calculated by the I-component rich constant IR (= IO + bIRCATSBFBA). Is gradually increased.
[0062]
Here, the I-component lean correction value bILCATSBFBA and the I-component rich correction value bIRCATSBFBA are set according to the degree of deterioration of the catalyst 10, and the deterioration of the catalyst 10 progresses, and the inversion cycle Ts of the output value RVO2 of the rear O2 sensor 13 Is gradually increased, the value is set to a larger value in an increasing function. Therefore, the output value FVO2 of the front O2 sensor 12 when the catalyst is new, which is indicated by a broken line in FIGS. , The gradient of the I-component rich constant IR and the I-component lean constant IL due to the integral control increases as indicated by the solid line.
[0063]
As a result, as the deterioration of the catalyst 10 progresses, the increase amount during the rich control and the decrease amount during the lean control become relatively large, whereby the air-fuel ratio fluctuation amount during the air-fuel ratio feedback control increases, To that extent, the O2 adsorption and desorption actions of the catalyst 10 are promoted, and it is possible to prevent deterioration of exhaust emissions due to progress of catalyst deterioration.
[0064]
The air-fuel ratio feedback correction coefficient LAMBDA is read in a fuel injection pulse width setting routine shown in FIG.
[0065]
This routine is executed at every predetermined crank angle (for example, 180 ° CA for a four-cylinder engine). First, at step S41, the engine speed Ne calculated based on the output signal from the crank angle sensor 14 and Based on the intake air amount Q detected by an intake air amount sensor (not shown) disposed downstream of the air cleaner (not shown), the basic fuel injection pulse width Tp is calculated from the following equation (8). I do.
Tp ← K · Q / NE (8)
Here, K is an injector characteristic correction constant.
[0066]
Next, in step S42, based on the coolant temperature, the throttle opening, and the like, various increase coefficients COEF related to the coolant temperature correction, the acceleration / deceleration correction, the full opening increase correction, and the like are set.
[0067]
Thereafter, the process proceeds to step S43, in which the air-fuel ratio feedback correction coefficient LAMBDA is read, and in step S44, the voltage correction pulse width Tv for interpolating the invalid injection time of the injector 7 by referring to the table with interpolation calculation based on the battery voltage VB. Set.
[0068]
Then, in step S45, the basic fuel injection pulse width Tp is air-fuel ratio corrected by the various increase coefficient COEF and the air-fuel ratio feedback correction coefficient LAMBDA, and the voltage is corrected by the voltage correction pulse width Tv to obtain the final fuel injection pulse width Ti. Set.
Ti ← Tp ・ COEF ・ LAMBDA + Tv (9)
[0069]
Then, in step S46, the fuel injection pulse width Ti is set in the injection timer of the fuel injection target cylinder, and the routine exits.
[0070]
As a result, the injection timer is started at a predetermined timing, a drive pulse signal having a fuel injection pulse width Ti is output to the injector 7 of the cylinder to be fuel-injected, and the predetermined fuel is injected from the injector 7.
[0071]
9 to 11 show a second embodiment of the present invention. In the first embodiment described above, the degree of deterioration of the catalyst 10 is determined based on the rich / lean inversion time of the rear O2 sensor 13, but in the present embodiment, the integral factor of the output value RVO2 of the rear O2 sensor 13 is determined. , The degree of deterioration of the catalyst 10 is determined.
[0072]
The catalyst deterioration diagnosis routine shown in FIG. 9 is performed only once or every predetermined time, for example, after the engine is started, after the catalyst 10 is activated, and when the engine operating state is in a steady state such as an idling operation. Is executed.
[0073]
First, in step S51, the current output value FVO2 of the front O2 sensor 12 is subtracted from the previous output value FVO2 (-1) of the front O2 sensor 12, and the absolute value of the change amount of the output value FVO2 of the front O2 sensor 12 ( FDVP is calculated (hereinafter referred to as “FVO2 change absolute value”) (FDVP ← | FVO2 (−1) −FVO2 |).
[0074]
In step S52, the current output value RVO2 of the rear O2 sensor 12 is subtracted from the previous output value RVO2 (-1) of the rear O2 sensor 12, and the absolute value of the change amount of the output value RVO2 of the rear O2 sensor 12 is obtained. RDVP is calculated (hereinafter referred to as “RVO2 change amount absolute value”) (RDVP ← | RVO2 (−1) −RVO2 |).
[0075]
Next, in step S53, the FVO2 change amount integrated value FDSVP is set by adding the FVO2 change amount absolute value FDVP to the integrated value of the FVO2 change amount absolute value FDVP (hereinafter, referred to as “FVO2 change amount integrated value”) FDSVP ( FDSVP ← FDSVP + FDVP). Note that the initial value of the FVO2 change amount integrated value FDSVP is 0.
[0076]
In step S54, the RVO2 change amount integrated value RDSVP is set by adding the RVO2 change amount absolute value RDVP to the integrated value of the RVO2 change amount absolute value RDVP (hereinafter, referred to as "RVO2 change amount integrated value") RDSVP ( RDSVP ← RDSVP + RDVP). Note that the initial value of the RVO2 change amount integrated value FDSVP is 0.
[0077]
Next, in step S55, it is checked whether or not a preset integration time has elapsed. If the integration time has not elapsed, the routine is exited as it is, and the calculation of the both change amount integration values FDSVP and RDSVP is continued. Do it.
[0078]
On the other hand, when the integration time has been reached, the process proceeds to step S56, and a comparison value CHANT between the RVO2 variation integration value RDSVP and the FVO2 variation integration value FDSVP is calculated (CHANT ← RDSVP / FDSVP).
[0079]
Then, in step S57, the table is referenced with interpolation calculation based on the comparison value CHANT, or the catalyst deterioration diagnosis value CATDIAGB is set by calculation.
[0080]
By the way, as the deterioration of the catalyst 10 progresses, the O2 storage effect of the catalyst decreases, and the waveform of the output value FVO2 of the rear O2 sensor 13 changes as shown in FIG. Since the waveform approximates, the RVO2 change amount integrated value RDSVP increases as the catalyst 10 deteriorates. As a result, the comparison value CHANT (= RDSVP / FDSVP) between the FVO2 change amount integrated value FDSVP and the RVO2 change amount integrated value RDSVP naturally increases as the deterioration of the catalyst 10 progresses.
[0081]
Therefore, it is possible to detect the degree of deterioration of the catalyst 10 from the comparison value CHANT. In the table referred to in step S57 described above, the larger the comparison value CHANT, the larger the catalyst deterioration diagnosis value CATDIAGB is stored. I have.
[0082]
Then, the process proceeds to step S58, where both the change amount integrated values FDSVP and RDSVP are cleared (FDSVP ← 0, RDSVP ← 0), and the routine exits.
[0083]
The catalyst deterioration diagnosis value CATDIAGB is read in a catalyst deterioration correction value setting routine shown in FIG. This routine is started every predetermined calculation cycle (32 msec in the present embodiment). First, in step S61, the value of the initialized flag FLGINI stored in the nonvolatile memory or the backup RAM (not shown) is changed. refer. This initialized flag FLGINI is reset (FLAGINI ← 0), for example, when data stored in the EPROM, the nonvolatile memory, or the backup RAM is not set, or is destroyed or lost.
[0084]
When FLGINI = 0, each set value data stored in the non-volatile memory or the backup RAM (not shown) has not been set, is destroyed, or has been lost. Therefore, the process proceeds to step S62 and proceeds to steps S62 to S64. Then, each set value stored in the nonvolatile memory or the backup RAM (not shown) is initialized.
[0085]
First, in step S62, the catalyst diagnosis learning value bCATDIAGB is initialized by a preset catalyst diagnosis initial value constant kCATDIAGB (bCATDIAGB ← kCATDIABB).
[0086]
In step S63, the I component rich correction value bIRCATSBBFBB is initialized with a preset I component rich correction initial value constant kIRCATSBBFBB (bIRCATSBBFBB ← kIRCATSBBFB).
[0087]
In step S64, the I minute lean correction value bILCATSBBFBB is initialized with a preset I minute lean correction initial value constant kILCATSBBFBB (bILCATSBBFBB ← kILCATSBBFBB).
[0088]
Note that the above-mentioned initial value constants kCATDIAGB, kIRCATSBFBB, and kILCATSBFBB are set to values corresponding to, for example, the catalyst 10 corresponding to a new product.
[0089]
Then, the process proceeds to step S65, where the initialization completion flag FLGINI is set (FLGINI ← 1), indicating that the initialization has been completed, and the routine exits.
[0090]
On the other hand, if it is determined in step S61 that the initialization of FLGINI = 1 has been set, the process proceeds to step S66 to read the catalyst deterioration diagnosis value CATDIAGB, and compares the catalyst deterioration diagnosis value CATDIAGB with the catalyst diagnosis learning value bCATDIAGB. . When CATDIAGB ≦ bCATDIAB, it is determined that the deterioration of the catalyst 10 has not progressed, and the routine exits from the routine.
[0091]
On the other hand, if CATDIAGB> bCATDIABB, it is determined that the catalyst 10 is deteriorating, and the process proceeds to step S67 to update the catalyst diagnosis learning value bCATDIAGB with the current catalyst deterioration diagnosis value CATDIABB (bCATDIAGGB ← CATDIAGB).
[0092]
Then, the process proceeds to step S68, based on the catalyst diagnosis learning value bCATDIABB updated in step S67, refers to the I component rich correction value setting table tIRCATSBBFBB shown in FIG. 11A with interpolation calculation, and obtains the I component rich correction value bIRCATSBFBBB. Set.
[0093]
In step S69, based on the catalyst diagnosis learning value bCATDIAGB updated in step S67, the I-minute lean correction value setting table tILCATSBBFBB shown in FIG. 11B is referenced with interpolation calculation, and the I-minute lean correction value bILCATSBBFBB is calculated. Set and exit the routine.
[0094]
As shown in FIGS. 11A and 11B, the I-component rich correction value setting table tIRCATSBFBB and the I-minute lean correction value setting table tILCATSBBFBB have the catalyst diagnosis learning value bCATDIAGB in relation to the catalyst diagnosis learning value bCATDIAGB. As the value increases, the I-rich correction value bIRCATSBBFBB and the I-lean correction value bILCATSBBFBB are stored.
[0095]
In the present embodiment, the correction values bIRCATSBBFBB and bILCATSBBFB stored in the I-rich correction value setting table tIRCATSBBFB and the I-lean correction value setting table tILCATSBBFBB show the same characteristics, but different characteristics. There may be.
[0096]
The initialization flag FLGINI, the I-rich correction value bIRCATSBBFBB, and the I-lean correction value bILCATSBBFBB are stored at predetermined addresses in a nonvolatile memory or a backup RAM, as in the first embodiment. The data is stored even after the switch is turned off, so that when the ignition switch is turned on next time, the data stored when the ignition switch was turned off last time can be read.
[0097]
The I component rich correction value bIRCATSBFBB and the I component lean correction value bILCATSBBFBB are used to correct the I component rich correction value bIRCATSBFBA to the I component rich correction value in the air-fuel ratio feedback correction coefficient setting routine shown in FIG. 4 of the first embodiment. The value bIRCATSBBFBB is replaced with the value, and the I-minute lean correction value bILCATSBFBA is replaced with the I-value lean correction value bILCATSBBFBB. Therefore, the description of the air-fuel ratio feedback correction coefficient setting routine here is omitted.
[0098]
FIGS. 12 to 15 show a third embodiment of the present invention. In the first embodiment described above, the I component rich constant IR and the I component lean constant IL of the air-fuel ratio feedback correction coefficient LAMBDA are corrected in accordance with the degree of catalyst deterioration. However, in the present embodiment, The P component rich constant PR and the P component lean constant PL are corrected according to the degree of catalyst deterioration.
[0099]
The catalyst deterioration correction value setting routine shown in FIG. 12 is applied instead of the catalyst deterioration correction value setting routine shown in FIG. 3 of the first embodiment.
[0100]
First, in step S71, the value of the initialized flag FLGINI stored in the nonvolatile memory or the backup RAM (not shown) is referred to. When FLGINI = 0, it is determined that each set value data stored in the non-volatile memory or the backup RAM (not shown) is unset, destroyed, or lost, and the process proceeds to step S72 and proceeds to step S72. In steps S74 to S74, each set value stored in the nonvolatile memory or the backup RAM (not shown) is initialized.
[0101]
First, in step S72, the catalyst diagnosis learning value bCATDIAGA is initialized with a preset catalyst diagnosis initial value constant kCATDIGA (bCATDIAGA ← kCATDIAGA).
[0102]
Further, in step S73, the P-component rich correction gain bPRCATGA is initialized with a preset P-component rich correction gain initial value constant kPRCATGA (bPRCATGA ← kPRCATGA).
[0103]
In step S74, the P-based lean correction gain bPLCATGA is initialized with a preset P-based lean correction gain initial value constant kPLCATGA (bPLCATGA ← kPLCATGA).
[0104]
The above-described initial value constants kCATDIGA, kPRCATGA, and kPLCATGA are set, for example, to values corresponding to the catalyst 10 when a new product is used.
[0105]
Then, the process proceeds to step S75, where the initialization flag FLGINI is set (FLGINI ← 1), indicating that the initialization is completed, and then the process jumps to step S80.
[0106]
On the other hand, if it is determined in step S71 that the initialization of FLGINI = 1 has been set, the process proceeds to step S76, in which the catalyst deterioration diagnosis value CATDIAGA set in the catalyst deterioration diagnosis routine shown in FIG. Then, the catalyst deterioration diagnosis value CATDIAGA is compared with the catalyst diagnosis learning value bCATDIAGA.
[0107]
Then, if CATDIAGA ≦ bCATDIAGA, it is determined that the deterioration of the catalyst 10 has not progressed, and the routine exits from the routine. On the other hand, when CATDIAGA> bCATDIAGA, it is determined that the deterioration of the catalyst 10 has progressed, and the process proceeds to step S77 to update the catalyst diagnosis learning value bCATDIAGA with the current catalyst deterioration diagnosis value CATDIAGA (bCATDIAGA ← CATDIAGA).
[0108]
Then, the process proceeds to step S78, and based on the catalyst diagnosis learning value bCATDIAGA updated in step S77, the P component rich correction gain setting table tPRCATGA shown in FIG. Set.
[0109]
Further, in step S79, based on the catalyst diagnosis learning value bCATDIAGA updated in step S77, the P-part lean correction gain setting table tPLCATGA shown in FIG. 14B is referenced with interpolation calculation, and the P-part lean correction gain bPLCATGA is obtained. After setting, the process proceeds to step S80.
[0110]
As shown in FIGS. 14A and 14B, the P-component rich correction gain setting table tPRCATGA and the P-component lean correction gain setting table tPLCATGA have the catalyst diagnosis learning value bCATDIAGA in relation to the catalyst diagnosis learning value bCATDIAGA. As P increases, the P-value rich correction gain bPRCATGA and the P-value lean correction gain bPLCATGA of larger values are stored.
[0111]
In the present embodiment, the correction values bPRCATGA and bPLCATGA stored in the P-rich correction gain setting table tPRCATGA and the P-lean correction gain setting table tPLCATGA respectively have the same characteristics, but have different characteristics. There may be.
[0112]
Further, the initialized flag FLGINI, the P-rich correction gain bPRCATGA, and the P-lean correction gain bPLCATGA are stored at predetermined addresses in a nonvolatile memory or a backup RAM, and the data is stored even after the ignition switch is turned off. When the ignition switch is turned on next time, the data stored when the ignition switch was turned off last time can be read.
[0113]
Then, when the process proceeds from step S75 or step S79 to step S80, the P-component rich basic correction value PRCATSBFB and the P-component rich correction gain bPRCATGA set in step S73 or S78 are read, and from the following equation (10), P The minute rich correction value PRCATSBFBA is calculated.
PRCATSBFBA ← PRCATSBFB · bPRCATGA (10)
Here, the P-component rich basic correction value PRCATSBFB is a fixed value in the present embodiment, but may be a variable value set according to the operating state of the engine.
[0114]
Next, in step S81, the P-part lean basic correction value PLCATTSBFB and the P-part lean correction gain bPLCATGA set in step S74 or S79 are read, and the P-part lean correction value PLCATSFBFBA is calculated from the following equation (11). Exit the routine.
PLCATSFBFBA ← PLCATSBFB · bPLCATGA (11)
Here, the P-based lean basic correction value PLCATSBFB is a fixed value in the present embodiment, but may be a variable value set according to the operating state of the engine.
[0115]
The P-rich correction value PRCATSBFBA and the P-lean correction value PLCATTBFBA are read in the air-fuel ratio feedback correction coefficient setting routine shown in FIG. This routine is applied instead of the air-fuel ratio feedback correction coefficient setting routine of the first embodiment shown in FIG.
[0116]
First, in step S91, it is checked whether the air-fuel ratio feedback control condition is satisfied. As the air-fuel ratio feedback control condition, for example, when the catalyst 10 is in an active state and the engine is in a steady operation state, it is determined that the air-fuel ratio feedback control condition is satisfied.
[0117]
When the air-fuel ratio feedback control condition is not satisfied, the process proceeds to step S92, and the air-fuel ratio feedback correction coefficient LAMBDA is set to 1.0 (LAMBDA ← 1.0). As a result, the air-fuel ratio control when the air-fuel ratio feedback control condition is not satisfied is open-loop control.
[0118]
Thereafter, the process proceeds to step S93, where the feedback control condition satisfaction flag FLGCLF is cleared (FLGCLF ← 0), and the routine exits.
[0119]
On the other hand, in step S91, it is determined that the air-fuel ratio feedback control condition is satisfied, and the process proceeds to step S94. Set LAMBDA.
[0120]
In step S94, the output value FVO2 of the front O2 sensor 12 is compared with the rich / lean inversion level FSL (see FIG. 7A) for judging the air-fuel ratio state, and when FVO2 ≧ FSL, the air-fuel ratio rich is satisfied. Then, the process proceeds to step S95, where the air-fuel ratio state at the time of the previous execution of the routine is checked. If the previous air-fuel ratio is lean, it is the first execution of the routine after the air-fuel ratio has been inverted from lean to rich, so that step S96 is executed. Then, the P-part lean correction value PLCATSFBFBA is read, and in the following step S97, the current air-fuel ratio feedback correction coefficient LAMBDA is calculated from the following equations (12) and (13).
LAMBDA ← LAMBDA-PL (12)
PL ← PO + PLCATSBFBA (13)
Here, PL is a P-part lean constant. PO is a basic proportional constant, and is a fixed value set in advance.
[0121]
As a result, as shown by the solid line in FIG. 15B, as the deterioration of the catalyst 10 progresses, the skip width of the P-part lean constant PL of the air-fuel ratio feedback correction coefficient LAMBDA is changed to the P-part lean correction value PLCATTSFBA for each calculation cycle. Is only gradually reduced.
[0122]
On the other hand, when it is determined in step S95 that the previous air-fuel ratio is lean, the rich state of the air-fuel ratio is continued, so the process proceeds to step S98, and the current air-fuel ratio feedback correction coefficient LAMBDA is set to the next value. It is calculated from equation (14).
LAMBDA ← LAMBDA-IL (14)
Here, IL is an I-minute lean constant, which is a preset fixed value.
[0123]
Then, the process proceeds from step S96 or S98 to step S99, sets the feedback control condition satisfaction flag FLGCLF (FLGCLF ← 1), and exits the routine.
[0124]
On the other hand, if it is determined in step S94 that the air-fuel ratio is lean with FVO2 <FSL, the flow branches to step S101 to check whether or not the air-fuel ratio at the time of execution of the previous routine is rich. Is the first routine execution after the air-fuel ratio is inverted from lean to rich, so the process proceeds to step S102, and the current air-fuel ratio feedback correction coefficient LAMBDA is calculated from the following equations (15) and (16). It returns to step S99.
LAMBDA ← LAMBDA + PR (15)
PR ← PO + PRCATSBFBA (16)
Here, PR is a P component rich constant.
[0125]
As a result, as shown by the solid line in FIG. 15A, as the degree of deterioration of the catalyst 10 progresses, the skip width of the P-component rich constant PR of the air-fuel ratio feedback correction coefficient LAMBDA increases with the P-component rich correction value for each calculation cycle. It is gradually increased by PRCATSBFBA.
[0126]
On the other hand, when it is determined in step S101 that the previous air-fuel ratio is lean, the process proceeds to step S104 because the lean state of the air-fuel ratio is continued, and the current air-fuel ratio is obtained from the following equation (17). The feedback correction coefficient LAMBDA is calculated, and the process returns to step S99.
LAMBDA ← LAMBDA + IR (17)
Here, IR is an I component rich constant, which is a fixed value set in advance.
[0127]
Therefore, the air-fuel ratio feedback correction coefficient LAMBDA is set by PI control according to the comparison result between the output value FVO2 of the front O2 sensor 12 and the rich / lean inversion level FSL, as shown in FIG.
[0128]
Then, a P-part lean constant PL during the P-part lean control in the PI control is set by adding the P-part lean correction value PLCATTSFBA to the basic proportional constant PO, and the air-fuel ratio is determined by the P-part lean constant PL (= PO + PLCATSBFBA). The feedback correction coefficient LAMBDA is skipped in the decreasing direction. Further, the P-rich rich constant PR at the time of the P-rich rich control is set by adding the P-rich rich correction value PRCATSBFBA to the basic proportional constant PO, and the air-fuel ratio feedback correction coefficient LAMBDA is obtained by the P-rich rich constant PR (= PO + PRCATSBBFBA). Is skipped in the increasing direction.
[0129]
Here, the P-part lean correction value PLCATSFBFBA and the P-part rich correction value PRCATSBFBA are set according to the degree of deterioration of the catalyst 10, and the deterioration of the catalyst 10 progresses, and the inversion cycle Ts of the output value RVO2 of the rear O2 sensor 13 15A, the output value FVO2 of the front O2 sensor 12 when the catalyst is new is indicated by a broken line in FIGS. 15A and 15B. As the time progresses, as shown by the solid line, the amount of change by the proportional control of the P-part rich constant PR and the P-part lean constant PL increases.
[0130]
As a result, as the deterioration of the catalyst 10 progresses, the increase amount during the rich control and the decrease amount during the lean control become relatively large, whereby the air-fuel ratio fluctuation amount during the air-fuel ratio feedback control increases, To that extent, the O2 adsorption and desorption actions of the catalyst 10 are promoted, and it is possible to prevent deterioration of exhaust emissions due to progress of catalyst deterioration.
[0131]
The air-fuel ratio feedback correction coefficient LAMBDA set in the air-fuel ratio feedback correction coefficient setting routine is read in the fuel injection pulse width setting routine shown in FIG. The fuel injection amount is corrected for the air-fuel ratio.
[0132]
16 and 17 show a fourth embodiment of the present invention. In the above-described second embodiment, the I component rich constant IR and the I component lean constant IL of the air-fuel ratio feedback correction coefficient LAMBDA are corrected in accordance with the degree of catalyst deterioration. The P component rich constant PR and the P component lean constant PL are corrected according to the degree of deterioration.
[0133]
The catalyst deterioration correction value setting routine shown in FIG. 16 is applied instead of the catalyst deterioration correction value setting routine shown in FIG. 10 of the second embodiment.
[0134]
In this routine, first, in step S111, the value of the initialized flag FLGINI stored in the nonvolatile memory or the backup RAM (not shown) is referred to. When FLGINI = 0, it is determined that each set value data stored in the non-volatile memory or the backup RAM (not shown) is unset, destroyed or lost, and the process proceeds to step S112 and proceeds to step S112. In steps S114 to S114, each set value stored in the nonvolatile memory or the backup RAM (not shown) is initialized.
[0135]
First, in step S112, the catalyst diagnosis learning value bCATDIAGB is initialized by a preset catalyst diagnosis initial value constant kCATDIAGB (bCATDIAGB ← kCATDIAGB).
[0136]
In step S113, the P-rich correction gain bPRCATGB is initialized with a preset P-rich rich gain initial value constant kPRCATGB (bPRCATGB ← kPRCATGB).
[0137]
In step S114, the P-lean correction gain bPLCATGB is initialized with a preset P-lean correction gain initial value constant kPLCATGB (bPLCATGB ← kPLCATGB).
[0138]
The above-mentioned initial value constants kCATDIAGB, kPRCATGB, and kPLCATGB are set to values corresponding to the catalyst 10 at the time of a new product, for example.
[0139]
Then, the process proceeds to step S115, in which the initialization flag FLGINI is set (FLGINI ← 1), indicating that the initialization has been completed, and then the process jumps to step S120.
[0140]
On the other hand, when it is determined in step S111 that the initialization of FLGINI = 1 has been set, the process proceeds to step S116, and the catalyst deterioration diagnosis value CATDIAGB set in the catalyst deterioration diagnosis routine shown in FIG. Then, the catalyst deterioration diagnostic value CATDIAGB is compared with the catalyst diagnostic learning value bCATDIAGB.
[0141]
When CATDIAGB ≦ bCATDIAB, it is determined that the deterioration of the catalyst 10 has not progressed, and the routine exits from the routine. On the other hand, when CATDIAGB> bCATDIABB, it is determined that the deterioration of the catalyst 10 has progressed, and the process proceeds to step S117 to update the catalyst diagnosis learning value bCATDIAGB with the current catalyst deterioration diagnosis value CATDIABB (bCATDIAGGB ← CATDIAGB).
[0142]
Then, the process proceeds to step S118, and based on the catalyst diagnosis learning value bCATDIAGB updated in step S117, the P-component rich correction gain setting table tPRCATGB shown in FIG. 17A is referred to with interpolation calculation to obtain the P-component rich correction gain bPRCATGB. Set.
[0143]
Also, in step S119, based on the catalyst diagnosis learning value bCATDIABG updated in step S117, the P-part lean correction gain setting table tPLCATGB shown in FIG. 17B is referenced with interpolation calculation, and the P-part lean correction gain bPLCATGB is calculated. After setting, the process proceeds to step S120.
[0144]
As shown in FIGS. 17A and 17B, the P-component rich correction gain setting table tPRCATGB and the P-component lean correction gain setting table tPLCATGB include the catalyst diagnosis learning value bCATDIAGGB in relation to the catalyst diagnosis learning value bCATDIAGB. As P increases, the P-rich correction gain bPRCATGB and the P-lean correction gain bPLCATGB of larger values are stored.
[0145]
In this embodiment, the correction values bPRCATGB and bPLCATGB stored in the P-rich correction gain setting table tPRCATGB and the P-lean correction gain setting table tPLCATGB, respectively, have the same characteristics, but have different characteristics. There may be.
[0146]
The initialized flag FLGINI, the P-rich correction gain bPRCATGB, and the P-lean correction gain bPLCATGB are stored at predetermined addresses in a non-volatile memory or a backup RAM, and the data is stored even after the ignition switch is turned off. When the ignition switch is turned on next time, the data stored when the ignition switch was turned off last time can be read.
[0147]
Then, when the process proceeds from step S115 or step S119 to step S120, the P-component rich basic correction value PRCATSBFB and the P-component rich correction gain bPRCATGB set in step S113 or S118 are read, and P is obtained from the following equation (18). The minute rich correction value PRCATSBBFBB is calculated.
PRCATSBFBB ← PRCATSBFB · bPRCATGB (18) Here, the P-component rich basic correction value PRCATSBFB is a fixed value in the present embodiment, but may be a variable value set according to the operating state of the engine.
[0148]
Next, in step S121, the P-part lean basic correction value PLCATSBFB and the P-part lean correction gain bPLCATGA set in step S114 or S119 are read, and the P-part lean correction value PLCATSBFBB is calculated from the following equation (19). Exit the routine.
PLCATTBFBBB ← PLCATSBFB · bPLCATGB (19)
Here, the P-based lean basic correction value PLCATSBFB is a fixed value in the present embodiment, but may be a variable value set according to the operating state of the engine.
[0149]
The P component rich correction value PRCATSBBFBB and the P component lean correction value PLCATTBFBB are read in the air-fuel ratio feedback correction coefficient setting routine shown in FIG. 13 of the third embodiment. In this case, the P-part lean correction value PLCATSFBFBA is replaced with the P-part rich correction value PRCATSBFBB, and the P-part rich correction value PRCATSBFBA is replaced with the P-part lean correction value PLCATTSFBB and applied.
[0150]
The air-fuel ratio feedback correction coefficient LAMBDA set in the air-fuel ratio feedback correction coefficient setting routine is read in the fuel injection pulse width setting routine shown in FIG. 5 of the first embodiment, and the air-fuel ratio feedback correction coefficient is set. The fuel injection amount is corrected for the air-fuel ratio by LAMBDA.
[0151]
As described above, in each of the above-described embodiments, according to the existing catalyst diagnosis logic, in the first embodiment and the third embodiment, the temporal element of the rich / lean inversion of the output value FVO2 of the rear O2 sensor 13 is used. In the second and fourth embodiments, the catalyst deterioration diagnostic value CATDIAGB is calculated from the integrated element of the output value FVO2 of the rear O2 sensor 13, and the catalyst deterioration diagnostic value CATDIAGB is calculated in the second and fourth embodiments. In the first embodiment, the deterioration diagnosis values CATDIAGA and CATDIAGB are adopted as parameters indicating the degree of deterioration of the catalyst 10, and in the air-fuel ratio feedback control performed by the PI control, based on the catalyst deterioration diagnosis value CATDIAGA or CATDIABB, Minute rich correction value bIRCATSBFBA, I minute lean correction value bILCATS FBA is set. In the second embodiment, the I-component rich correction value bIRCATSBFBB and the I-component lean correction value bILCATSBFBB are set. In the third embodiment, the P-component rich correction gain bPRCATGA and the P-component lean correction gain ILCATGA are set. In the fourth embodiment, the P-rich correction gain bPRCATGB and the P-lean correction gain bPLCATGB are set, so that a highly efficient O2 storage effect according to the degree of deterioration of the catalyst 10 is always achieved. Can be realized.
[0152]
As a result, for example, when the catalyst 10 is new, in the first embodiment and the third embodiment, a small P-part rich constant PR and a small P-part lean constant PL are set, and in the second embodiment, In the embodiment and the fourth embodiment, since the I component rich constant IR and the I component lean constant IL are set to small values, the fuel reduction amount and the fuel injection amount correction amount by the air-fuel ratio feedback control are set to be small. In addition, it is possible to prevent the air-fuel ratio from being excessively corrected by the air-fuel ratio feedback, and to reduce the exhaust emission.
[0153]
In a situation where the O2 storage effect of the catalyst decreases as the deterioration of the catalyst 10 progresses, the P-rich constant PR and the P-lean constant PL are set to correspondingly large values. The fuel reduction correction and the fuel increase correction are increased by the correction, and the O2 adsorption and desorption of the catalyst are promoted by the increase in the air-fuel ratio fluctuation amount by the air-fuel ratio feedback control, so that the activation of the catalyst can be realized. Even if the deterioration proceeds, it is possible not only to prevent deterioration of the exhaust emission, but also to extend the functional life of the catalyst.
[0154]
Further, since the air-fuel ratio feedback correction can be optimally performed in accordance with the state of the catalyst, even if the catalyst has a performance lower than the O2 storage effect of the existing catalyst, the same O2 storage effect as the existing catalyst is obtained. It is possible to reduce the production cost of the catalyst.
[0155]
The present invention is not limited to the above-described embodiments. For example, the air-fuel ratio feedback correction coefficient LAMBDA is calculated based on the catalyst deterioration diagnostic values CATDIAGA and CATDIAGB based on the I component constants IR and IL and the P component constants PR and PL. May be corrected. It is not necessary to limit the number of catalysts, and a plurality of catalysts may be provided between the front O2 sensor 12 and the rear O2 sensor 13.
[0156]
【The invention's effect】
As described above, according to the present invention, the air-fuel ratio feedback correction is performed in accordance with the degree of deterioration of the catalyst, so that not only when the catalyst is new, but also when the catalyst is deteriorated, the exhaust gas is exhausted. It is possible to prevent deterioration of the emission, and it is possible to efficiently utilize the O2 storage effect of the catalyst.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of an engine control system according to a first embodiment.
FIG. 2 is a flowchart showing a catalyst deterioration diagnosis routine;
FIG. 3 is a flowchart showing a catalyst deterioration correction value setting routine of the same.
FIG. 4 is a flowchart showing an air-fuel ratio feedback correction coefficient setting routine.
FIG. 5 is a flowchart showing a fuel injection pulse width setting routine;
6A is an explanatory diagram of an I component rich correction value setting table, and FIG. 6B is an explanatory diagram of an I component lean correction value setting table.
FIG. 7 is a time chart showing the relationship between the output value of the front O2 sensor and the air-fuel ratio feedback correction coefficient.
8A is an explanatory diagram of an I-component rich correction value of an air-fuel ratio feedback correction coefficient, FIG. 8B is an explanatory diagram of an I-component lean correction value of an air-fuel ratio feedback correction coefficient, and FIG. ) Is an enlarged view of a part c, and (d) is an enlarged view of a d part of (b).
FIG. 9 is a flowchart illustrating a catalyst deterioration diagnosis routine according to a second embodiment.
FIG. 10 is a flowchart showing a catalyst deterioration correction value setting routine of the same.
11A is an explanatory diagram of an I component rich correction value setting table, and FIG. 11B is an explanatory diagram of an I component lean correction value setting table.
FIG. 12 is an explanatory diagram of a catalyst deterioration correction value setting routine according to a third embodiment.
FIG. 13 is a flowchart showing an air-fuel ratio feedback correction coefficient setting routine;
14A is an explanatory diagram of a P-component rich correction gain setting table; FIG. 14B is an explanatory diagram of a P-component lean correction gain setting table;
FIG. 15A is an explanatory diagram of a P-component rich correction value of an air-fuel ratio feedback correction coefficient, and FIG. 15B is an explanatory diagram of a P-component lean correction value of an air-fuel ratio feedback correction coefficient;
FIG. 16 is a flowchart illustrating a catalyst deterioration correction value setting routine according to a fourth embodiment.
17A is an explanatory diagram of a P-component rich correction gain setting table, and FIG. 17B is an explanatory diagram of a P-component lean correction gain setting table.
FIG. 18 is a time chart showing a relationship between a conventional front O2 sensor output voltage, a rear O2 sensor output voltage, and an air-fuel ratio feedback correction coefficient.
[Explanation of symbols]
1 engine
10 Catalyst
12 Front O2 sensor (Front air-fuel ratio sensor)
13 Rear O2 sensor (Rear air-fuel ratio sensor)
CATDIAGA, CATDIAGB Catalyst deterioration diagnosis value
CHANT comparison value
FDSVP FVO2 change amount integrated value
FDVP FVO2 change absolute value
FVO2, RVO2 output value (output voltage)
bILCATSBFBA, bILCATSBFBB I-minute lean correction value
bIRCATSBFBA, bIRCATSBFBB I component rich correction value
bPLCATGA, bPLCATGB Lean correction gain for P
bPRCATGA, bPRCATGB P-component rich correction gain
IL I minute lean constant
IR I component rich constant
LAMBDA air-fuel ratio feedback correction coefficient
PLP lean constant
PLCATSFBFBA, PLCATSFFBBB P-part lean correction value
PRP component rich constant
PRCATSBFBA, PRCATSBFBB P-component rich correction value
RDSVP RVO2 change amount integrated value
RDVP RVO2 change absolute value
Ti fuel injection pulse width (fuel injection amount)
Ts inversion cycle

Claims (4)

A front air-fuel ratio sensor and a rear air-fuel ratio sensor are respectively disposed upstream and downstream of the catalyst interposed in the exhaust system of the engine,
An air-fuel ratio feedback control device for an engine that sets an air-fuel ratio feedback correction coefficient by proportional integral control according to the output values of the two air-fuel ratio sensors and corrects the fuel injection amount with the air-fuel ratio feedback correction coefficient.
Reversal cycle calculating means for detecting a rich / lean reversal cycle of the output value of the rear air-fuel ratio sensor;
Catalyst deterioration diagnosis value setting means for setting a catalyst deterioration diagnosis value based on the inversion cycle,
Correction means for setting a correction value for correcting at least one of a proportional constant and an integral constant of the air-fuel ratio feedback correction coefficient based on the catalyst deterioration diagnosis value;
An air-fuel ratio feedback control device for an engine, comprising: A front air-fuel ratio sensor and a rear air-fuel ratio sensor are respectively disposed upstream and downstream of the catalyst interposed in the exhaust system of the engine,
An air-fuel ratio feedback control device for an engine that sets an air-fuel ratio feedback correction coefficient by proportional integral control according to the output values of the two air-fuel ratio sensors and corrects the fuel injection amount with the air-fuel ratio feedback correction coefficient.
A front air-fuel ratio sensor output value change amount calculating means for calculating a change amount of the front air-fuel ratio sensor output value,
A rear air-fuel ratio sensor output value change amount calculating means for calculating a change amount of the rear air-fuel ratio sensor output value,
Comparison value calculation means for calculating a comparison value between the change amount of the front air-fuel ratio sensor output value and the change amount of the rear air-fuel ratio sensor output value,
Catalyst deterioration diagnosis value setting means for setting a catalyst deterioration diagnosis value based on the comparison value;
Correction means for setting a correction value for correcting at least one of a proportional constant and an integral constant of the air-fuel ratio feedback correction coefficient based on the catalyst deterioration diagnosis value;
An air-fuel ratio feedback control device for an engine, comprising: The correction means sets an integral rich correction value and an integral lean correction value based on the catalyst deterioration diagnosis value,
3. The air-fuel ratio feedback control device for an engine according to claim 1, wherein the air-fuel ratio feedback correction coefficient is corrected by the two correction values. The correcting means sets a proportional rich correction gain and a proportional lean correction gain based on the catalyst deterioration diagnosis value,
A proportional rich correction value and a proportional lean correction value are set based on the two correction gains,
3. The air-fuel ratio feedback control device for an engine according to claim 1, wherein the air-fuel ratio feedback correction coefficient is corrected by the two correction values.

Images