JP2853566B2 - Engine catalyst deterioration diagnosis device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for diagnosing catalyst deterioration of an engine.

[0002]

The O ₂ sensor disposed in the downstream side as well upstream of the Prior Art catalyst (three-way catalyst), and executes a feedback control of the air-fuel ratio based on the output of the upstream O ₂ sensor, both O ₂ There is a device for diagnosing catalyst deterioration by comparing the outputs of sensors (for example, see Japanese Patent Application Laid-Open No. 63-2).
05441).

To explain this, during feedback control of the air-fuel ratio based on the output of the upstream O ₂ sensor,
The output of the upstream O ₂ sensor periodically repeats rich and lean as shown in FIG. 15 (a), while the downstream oxygen concentration gradually changes due to the O ₂ storage capacity of the catalyst downstream of the catalyst. since the objects, the output of the downstream O ₂ sensor, the period becomes longer than the upstream O ₂ sensor as in FIG. 15 (b), it takes a substantially constant value without fluctuation range.

In this case, if the catalyst deteriorates,
O ₂ the change in the oxygen concentration between the upstream and the downstream side of the catalyst is not so different in reduction in the storage capacity, the output of the downstream O ₂ sensor, as shown in FIG. 15 (c), the upstream O ₂ sensor Inversion is repeated at a cycle approximating the output of. Rich deterioration of the catalyst downstream O ₂ sensor output is not appearing in the inversion period of the lean.

Accordingly, the upstream _{O 2} sensor output rich, lean inversion period T1 and the downstream _{O 2} sensor rich, when determining the ratio between the lean inversion period T2 (T1 / T2), the ratio (T1 / T2) is a value close to 0 when the catalyst is new, but approaches 1 as the deterioration of the catalyst progresses. Therefore, this ratio (T1 / T2) is determined as a reference value (for example, 0.5). When this is the case, it can be determined that the catalyst has deteriorated.

[0006]

However, there is a certain degree of deterioration of the catalyst, and when the catalyst is roughly divided into three categories: small deterioration, medium deterioration, and large deterioration, the catalyst cannot sufficiently purify the exhaust gas at the degree of medium deterioration. In this case, there is no need for diagnosing deterioration, and diagnosing deterioration at a moderate deterioration level wastes resources.

However, in the above device, during the air-fuel ratio feedback control centered on the stoichiometric air-fuel ratio, the output of the downstream O ₂ sensor is most likely to be repeatedly inverted. The output of the O ₂ sensor is approximately similar, and the output of the downstream O ₂ sensor varies, so that it is impossible to separate whether the catalyst is in moderate deterioration or large deterioration.

[0008] where, performs the deterioration diagnosis of the catalyst based on the air-fuel ratio from the stoichiometric air-fuel ratio in the output of both sensors in a state of being shifted to the lean side or the rich side, the exhaust purifying the catalyst to the extent of the middle deterioration capacity It is conceivable that the catalyst is not deemed to be deteriorated when it still remains, so that the catalyst can be operated without waste until immediately before it is determined to be greatly degraded.

[0009] However, even for the diagnosis of deterioration,
Even when the medium is new, the stoichiometric air-fuel ratio is set to lean or rich.
Shifting to the side has a disadvantage in terms of exhaust performance.
That the at until cause performance degradation over middle deteriorated catalyst
Does not shift the air-fuel ratio toward lean or rich
Avoid shifting the air-fuel ratio from the stoichiometric air-fuel ratio when new
Can be

On the other hand, when the average air-fuel ratio is close to the stoichiometric air-fuel ratio
Although it is not maintained with high accuracy,
Therefore, for example, performing a lean shift of the air-fuel ratio
Lean shift amount becomes unstable, and medium and large deterioration are separated
Accuracy deteriorates.

Therefore, the present invention provides a method for reducing the degree of deterioration to a small degree
The determination of the catalyst is made under the stoichiometric air-fuel ratio first,
Exhaust due to air-fuel ratio deviating from stoichiometric air-fuel ratio
Prevent disadvantages in terms of performance
Then, the judgment of medium deterioration and large deterioration is outside the stoichiometric air-fuel ratio.
The catalyst is judged to be greatly deteriorated by performing at the adjusted air-fuel ratio.
Work just before you learn
Only during the operation, the judgment of medium deterioration and large deterioration
By stabilizing the amount deviated from the stoichiometric air-fuel ratio
The purpose is to improve the separation accuracy between degradation and major degradation
You.

[0012]

According to a first aspect of the present invention, as shown in FIG. 16 , the basic injection amount Tp is changed according to the operating conditions of the engine.
, A sensor (for example, an O ₂ sensor or an air-fuel ratio sensor) 22 that outputs according to the oxygen concentration in the exhaust gas on the upstream side of the catalyst, and a feedback constant (for example, Means 23 for calculating a proportional component or an integral component), means 41 for storing a learning value for the feedback constant, means for correcting the feedback constant with the learning value read from the storage means 24, and the corrected Means 43 for calculating an air-fuel ratio feedback correction amount α using a feedback constant, and means 2 for calculating the fuel injection amount by correcting the basic injection amount Tp with the air-fuel ratio feedback correction amount α.
5 and means 26 for supplying this amount of fuel to the intake pipe
And a sensor 27 for outputting an output corresponding to the oxygen concentration in the exhaust gas on the downstream side of the catalyst, and learning of the storage means 41 in a direction for returning the air-fuel ratio to the stoichiometric air-fuel ratio based on the output of the downstream sensor 27. Means 43 for updating the value; means 44 for determining whether or not the number of updates by the updating means 43 is greater than or equal to a predetermined number; and when the number of updates is greater than or equal to the predetermined number from this determination result, the two sensor outputs are compared. Means 3 for determining whether small deterioration or medium deterioration or more has occurred in the catalyst
1 and when the catalyst shows moderate deterioration or more based on this determination result, the updating of the learning value by the updating means 43 is prohibited and the air-fuel ratio is deviated from the stoichiometric air-fuel ratio (either lean or rich). Means 45 for changing the learning value read from the storage means 41 to a learning value for diagnosis, and comparing the two sensor outputs in an air-fuel ratio state deviated from the stoichiometric air-fuel ratio by the learning value for diagnosis, and And means 29 for determining whether large deterioration or medium deterioration has occurred.

According to a second aspect of the present invention, as shown in FIG. 17 , means 2 for calculating a basic injection amount Tp according to the operating conditions of the engine.
1 and a sensor (for example, an O ₂ sensor or an air-fuel ratio sensor) that outputs according to the oxygen concentration in the exhaust gas upstream of the catalyst
22, means 23 for calculating a feedback constant (for example, a proportional component or an integral component) based on the output of the upstream sensor 22, and each learning area divided into a plurality of areas using at least the engine speed and the load as parameters. Means 51 for independently storing a learning value for the feedback constant, and means 52 for reading a learning value of a learning area to which the engine speed and load belong at that time.
Means 42 for correcting the feedback constant with the read learning value; means 24 for calculating the air-fuel ratio feedback correction amount α using the corrected feedback constant; Means 2 for calculating fuel injection amount by correcting injection amount Tp
5 and means 26 for supplying this amount of fuel to the intake pipe
A sensor 27 that outputs an output corresponding to the oxygen concentration in the exhaust gas downstream of the catalyst; and a learning value of the same learning area when the engine speed and load at that time remain in the same learning area for a predetermined period. 53 for updating the air-fuel ratio to the stoichiometric air-fuel ratio side based on the output of the downstream sensor 27, and whether the number of updates in the learning area overlapping the diagnostic area by the updating means 53 is equal to or greater than a predetermined number. Means 54 for determining whether the catalyst has undergone a small deterioration or a medium deterioration or more in the learning area overlapping the diagnosis area by comparing the two sensor outputs when the number of updates is equal to or more than the predetermined number of times. The determining means 31 and a learning area overlapping with the diagnostic area by the updating means 53 when the catalyst has a medium deterioration or more in a learning area overlapping the diagnostic area based on the determination result. The learning value read from the learning area overlapping the diagnosis area by the storage means 51 in a direction in which the air-fuel ratio deviates from the stoichiometric air-fuel ratio (either lean or rich) is prohibited. Means 55 for changing to a learning value, and comparing the two sensor outputs in an air-fuel ratio state deviating from the stoichiometric air-fuel ratio based on the learning value for diagnosis, to determine whether the catalyst has undergone large deterioration or medium deterioration. Means 29 are provided.

A third invention is the first or your <br/> are in the second inventions, as shown in FIG. 18, one of the determining means 29 the which of the large deterioration and medium deterioration occurs, each Means 61, 62 for calculating the number of reversals or reversal periods from sensor outputs
A means 63 for calculating the ratio or difference of the number of inversions or the inversion cycle, a means 64 for calculating a weighted average of the ratio or the difference, and a means 65 for comparing the weighted average with the determination reference value. .

[0015]

When the average air-fuel ratio is close to the stoichiometric air-fuel ratio by the air-fuel ratio feedback control, the upstream side and the downstream side differ depending on whether the catalyst has advanced to a large degree of deterioration or when the catalyst has advanced only to the degree of moderate deterioration. Since the two sensors have substantially the same output (in the case of the O ₂ sensor, both of the large deterioration and the medium deterioration are rich and the reversal period of the lean is almost the same), it is difficult to separate the medium deterioration from the large deterioration. Although it is difficult, when the air-fuel ratio is deviated from the stoichiometric air-fuel ratio due to the feedback constant for diagnosis in the first invention, the output of the downstream sensor differs between the case of moderate deterioration and the case of large deterioration (theoretical theory). when O ₂ sensor that will output the air-fuel ratio as substantially the center is reversed, rich output of the downstream side of the O ₂ sensor when the large degradation, though there is no change in the inversion cycle of a lean, downstream in the medium deteriorate O ₂ sensor output rich, the inversion period of the lean becomes longer). That is, at the time of catalyst deterioration diagnosis, by forcibly removing the air-fuel ratio from the stoichiometric air-fuel ratio, the output of the sensor on the downstream side becomes different, so that it is possible to separate medium deterioration from large deterioration.

In this way, when it is possible to separate the medium deterioration from the large deterioration, the deterioration is diagnosed only when the deterioration progresses, and the deterioration is not diagnosed when the degree of the medium deterioration still has the purifying ability. This allows the catalyst to work to its fullest extent just prior to proceeding to major degradation.

Further, when the performance of the catalyst deteriorates more than the degree of moderate deterioration, even if the air-fuel ratio is lean-shifted for diagnosis, the exhaust performance does not extremely decrease.

In the first invention, before performing the deterioration diagnosis with the air-fuel ratio deviated from the stoichiometric air-fuel ratio, it is determined from the comparison between the two sensor outputs whether the catalyst has undergone small deterioration or medium deterioration or more. When the deterioration diagnosis is performed by deviating the air-fuel ratio from the stoichiometric air-fuel ratio only when medium degradation or higher occurs, the air-fuel ratio is not deviated from the stoichiometric air-fuel ratio when the catalyst is new,
As a result, there is no disadvantage in terms of exhaust performance.

In the first aspect of the invention, the learning value of the storage means 41 is updated in a direction to return the air-fuel ratio to the stoichiometric air-fuel ratio based on the sensor output on the downstream side of the catalyst, and the number of times of updating becomes greater than a predetermined number. When the deterioration diagnosis is performed with the air-fuel ratio deviated from the stoichiometric air-fuel ratio, the air-fuel ratio accompanying the unevenness of the output of the upstream sensor and the non-equilibrium gas on the upstream side of the catalyst is detected at the timing when the number of updates exceeds a predetermined number. Regardless of the variation, the air-fuel ratio changes periodically around the stoichiometric air-fuel ratio regardless of these factors. The separation accuracy of deterioration is improved.

In the second invention, a learning value for a feedback constant is stored independently for each learning area divided into a plurality of areas using at least the engine speed and the load as parameters, and the engine speed and the load at that time are stored. While the value remains in the same learning area for a predetermined period of time, the learning value of the same learning area is updated based on the output of the downstream sensor 27 so as to return the air-fuel ratio to the stoichiometric air-fuel ratio, while the diagnostic area is updated. If the deterioration diagnosis is performed with the air-fuel ratio deviated from the stoichiometric air-fuel ratio when the number of updates in the learning area that overlaps with the predetermined number of times is greater than or equal to the stoichiometric air-fuel ratio, the accuracy of the learning value becomes better than in the first invention, and the medium deterioration And the separation accuracy of large deterioration is improved. If the learning area can be subdivided, the control accuracy to the stoichiometric air-fuel ratio is improved, and the air-fuel ratio in the diagnosis area is closer to the stoichiometric air-fuel ratio. Separation accuracy) is also improved.

[0021] In the third invention, the first or your <br/> are in the second inventions, the the weighted average value of the determination which of the large deterioration and medium deterioration occurs is used, the degradation and the medium further The separation accuracy of large deterioration is improved.

[0022]

1, reference numeral 1 denotes an engine main body, 2 denotes an intake passage, 3 denotes an exhaust passage, 4 denotes a throttle valve, 5 denotes a fuel injector, and 6 denotes a catalyst. Fuel regulated to a constant pressure is supplied to the fuel injector 5 via a fuel supply system (not shown), and the fuel injector 5 is opened by a drive pulse from the control unit 15 and is proportional to the valve opening pulse width. An amount of fuel is injected.

Reference numeral 11 denotes a hot-wire type air flow meter for detecting an air flow rate, and 12 denotes a crank angle reference position (4
A crank angle sensor that outputs a signal (Ref signal) every 180 ° for a cylinder and every 120 ° for a 6 cylinder) and a signal for each unit crank angle. Reference numerals 13 and 14 denote stoichiometric air-fuel ratios according to the residual oxygen concentration in the exhaust gas. Output changes suddenly after
Reference numeral ₂ denotes a sensor, which is a water temperature sensor for detecting the temperature of the cooling water of the engine. A control unit 15 to which signals from these sensors are input performs air-fuel ratio control so that the average air-fuel ratio becomes the stoichiometric air-fuel ratio.

The air-fuel ratio control in the control unit 15 comprising a microcomputer is as follows.

The fuel injector 5 is driven in synchronization with the Ref signal. For example, in the simultaneous injection system, the engine 1
Once every rotation, simultaneously for all cylinders Ti = Te + Ts (1) where Te: effective pulse width Ts: injection pulse width Ti given by the formula of invalid pulse width according to battery voltage
Is activated, and when the sequential injection method is adopted, the injector 5 has an injection pulse width Ti given by the formula of Ti = 2 × Te + Ts (2) once for every two revolutions of the engine for each cylinder.
Is activated.

FIG. 2 is a flowchart for calculating the above Ti, which is executed at a constant period (for example, 10 msec).

In step 1, the basic pulse width Tp is calculated from the air flow rate Q detected by the air flow meter 11 and the engine speed N detected by the crank angle sensor, as follows: Tp = (Q / N) × K (3) where K: Calculate with the constant formula. The air-fuel ratio determined by this Tp is called the base air-fuel ratio.

In step 2, the effective pulse width Te in the above equations (1) and (2) is calculated by using the basic pulse width Tp as follows: Te = Tp × α × αmxCo (4) where α: air-fuel ratio feedback correction coefficient αm : Basic air-fuel ratio learning value Co: Various correction coefficients Equations are calculated, and in step 3, different values of Ti are calculated depending on whether all cylinders are simultaneously or sequentially injected.

The various correction coefficients Co in the equation (4) are values for ensuring smooth operation under various conditions. For example, the basic pulse width Tp is corrected based on a signal from each sensor such as a water temperature sensor at the time of starting, warming up, high load, and the like. At this time, the value of the air-fuel ratio feedback correction coefficient α described later is clamped at 100% (steps 11 and 12 in FIG. 3).

The air-fuel ratio feedback correction coefficient α in the equation (4) is calculated by proportional integral control (a type of feedback control) based on the output of the O ₂ sensor 13 on the upstream side of the catalyst 6.
The value obtained in synchronization with the ef signal, and the value of α is 100%
Is exceeded, the air-fuel ratio is shifted to the rich side from equation (3),
When the value falls below, the air-fuel ratio is returned to the lean side.

FIG. 3 is a flowchart for calculating the air-fuel ratio feedback correction coefficient α, which is executed in synchronization with the Ref signal.

Since the output of the upstream O ₂ sensor is high (about IV) on the rich side and low (substantially 0 V) on the lean side of the stoichiometric air-fuel ratio, it exceeds the slice level provided per 0.5 V. Therefore, when the actual air-fuel ratio is on the rich side, and when the actual air-fuel ratio is smaller than the slice level, the air-fuel ratio is on the lean side. Therefore, the upstream O ₂ sensor output (indicated by simply "front O _2" in the figure), for example, when inverted from the rich side to the lean side, the value obtained by adding the proportional amount P _L to the previous feedback correction coefficient α Is updated as the current α (steps 13, 15, 22, and
25), the _{O 2} sensor output from the next time adds integrated amount _{I L} until immediately before reversal to the rich side (step in FIG. 3 13,15,26,27), the injection quantity by the fuel increase by alpha (injection pulse width Ti) increases and the actual air-fuel ratio gradually increases.

[0033] Consequently, when the O ₂ sensor output is inverted to the rich side, it updates the value obtained by subtracting the proportional part P _R from the last alpha as the current alpha (Step of Fig. 3 13, 1
4,16,19), from the next time subtracts the integrated amount _{I R} until just before _{the O 2} sensor output is inverted again to the lean side (step 13,14,20,21 in Figure 3).

[0034] When the O ₂ sensor output is inverted to the lean side repeated, as described above.

By such repetition, the actual air-fuel ratio changes at a predetermined cycle, and the average air-fuel ratio is maintained within a window (a predetermined air-fuel ratio range around the stoichiometric air-fuel ratio). It is.

Steps 25 and 19 not described.
For the learning value LP of (see steps 17 and 1
8, 23, and 24) will be described below.

Since the unburned components in the exhaust gas have been reacted by the catalyst 6 downstream of the catalyst 6, unlike the upstream side of the catalyst 6, stable measurement of equilibrium gas is obtained for the O ₂ sensor. Output is obtained. When executing the air-fuel ratio feedback control based on the upstream O ₂ sensor 13, the air-fuel ratio due to the effect of non-equilibrium gas on the upstream side of the catalyst is downstream O ₂ sensor output if the rich tendency of substantially constant richer value, also when the air-fuel ratio is reversed lean tendency downstream O ₂ sensor output is not indicate a substantially constant value in the lean side.

[0038] Using these downstream O ₂ sensor output, when the downstream O ₂ sensor output is indicative of a rich trend, Yari back the air-fuel ratio to the lean side, and when the downstream O ₂ sensor indicates a lean trend If the air-fuel ratio is returned to the rich side, the average air-fuel ratio can be more accurately controlled to the stoichiometric air-fuel ratio.

For example, as shown in FIG. 3, the proportional value P _{L is obtained} by the learning value LP for the proportional component at the time of inversion to the lean side.
And (step 25 of FIG. _{3) P L = P L +} LP ... by formula (5), wherein the proportional amount _{P R} at the time of reversal to the rich side to the opposite P _R = _P R -LP ... (6) (Step 19 in FIG. 3).

The learning values LP in the equations (5) and (6) are read out for each learning area shown in FIG. In FIG. 4 (subroutine of steps 18 and 24 in FIG. 3), step 31
Then, the learning area to which the engine speed N and the basic pulse width (engine load equivalent amount) Tp at that time belongs is determined, and the learning value LP stored in that area is read in step 32. Steps 33 and 34 in FIG. 4 will be described later.

On the other hand, the learning value LP is updated when the operating conditions of the engine have been in the same learning area for a certain period.

FIG. 6 (a subroutine of steps 17 and 23 in FIG. 3) is a flowchart showing the updating of the learning value LP.
_It is executed every time the _two sensor outputs (or the air-fuel ratio feedback correction coefficient α) are inverted.

[0043] In FIG. 6, at a portion from step 41 to step 45 see whether the update condition, the following conditions <1> that is a feedback control region of the downstream O ₂ sensor (step 41), <2> the rotational speed n and the basic pulse width Tp are in the same learning area (step 43), <3> the counter value _{j R} exceeds a predetermined value _{n R} and within the same learning area (step 44 , 45), when all of the above are satisfied, it is determined that the update condition is satisfied, and the routine proceeds to step 47. Step 46 will be described later.

[0044] If the downstream _{O 2} sensor output is on the rich side in step 47, LP = LP-DLPR ... (7) However, DLPR: learning value LP of the learning area that has been established for update conditions in the formula of a constant value
Rewrite the smaller side is re-stored in the same learning area (step 49), when the downstream _{O 2} sensor output is lean in the opposite, LP = LP + DLPL ... ( 8) However, DLPL: constant value (Steps 50 and 49).

The learning value LP is stored independently for each learning area, and is backed up by a vehicle-mounted battery so that the value is not lost even after the engine is stopped.

As described above, by using the output of the downstream O ₂ sensor also during the air-fuel ratio control, the output variation of the upstream O ₂ sensor and the variation of the air-fuel ratio due to the non-equilibrium gas upstream of the catalyst are reduced. However, regardless of these, the air-fuel ratio periodically changes around the stoichiometric air-fuel ratio.

[0047] The above (5), (6) is proportional portion _P L, _P is an example of introducing a learned value for the _R, integral constant _I L Alternatively or in addition, for the _{I R} Learning values can also be introduced.

By the way, when the catalyst deteriorates, O _{2 of the} catalyst is deteriorated.
Storage capacity downstream O ₂ sensor output rich in reduction of lean inversion period (or number of reversals) is shortened, the upstream O ₂ sensor output rich, so approaches the inversion period of the lean, the two O ₂ It is possible to diagnose whether the catalyst has deteriorated based on the ratio (or difference) of the rich and lean reversal periods of the sensor output. However, during the air-fuel ratio feedback control around the stoichiometric air-fuel ratio, the downstream O _{Since the} output of the _two sensors is most likely to be repeatedly inverted, the output of the downstream O ₂ sensor is approximately similar between the medium deterioration and the large deterioration, and the output of the downstream O ₂ sensor also varies. It cannot be separated into medium and large deterioration.

To cope with this, the control unit 15 forcibly shifts the air-fuel ratio from the stoichiometric air-fuel ratio to the lean side (or the rich side) at the time of the deterioration diagnosis, and the two sensors in the shifted air-fuel ratio state. It is determined whether the catalyst is in the state of large deterioration or medium deterioration based on the output of .If the degree of the medium deterioration still has the exhaust purification ability of the catalyst, the warning light 16 provided on the operation panel or the like is turned on. The deterioration diagnosis is ended without performing, and the warning lamp 16 is turned on only when the deterioration becomes large.

First, the reason why the air-fuel ratio is forcibly shifted from the stoichiometric air-fuel ratio to the lean side and the catalyst deterioration is diagnosed based on the outputs of both sensors in the shifted state is as follows. .

As shown in FIG. 7, for a moderately degraded catalyst,
Since the O ₂ storage capacity still remains, the oxygen concentration decreases stepwise on the rich side from the stoichiometric air-fuel ratio (0
On the lean side, the CO concentration decreases stepwise (approaches 0), whereas when the deterioration further progresses to a large deterioration state, the O ₂ storage capacity almost disappears. The changes in the oxygen concentration and the CO concentration are almost the same as those in the upstream side of the catalyst. Such a difference in both the O ₂ storage capability, appears when brought into slightly shifted to the lean side from the stoichiometric air-fuel ratio, as shown in FIG. 8, only when the medium degradation downstream O ₂
Sensor output is less likely to reverse.

Therefore, when the air-fuel ratio was shifted from the stoichiometric air-fuel ratio to the lean side, an experiment was conducted to determine how the reversal frequency ratio HZRATE changes between medium deterioration and large deterioration.
Are obtained. From FIG. 9, near the stoichiometric air-fuel ratio, the value of the reversal frequency ratio HZRATE is close between large deterioration and medium deterioration, so that it was difficult to separate them. It is easier to do. That is, by forcibly shifting the air-fuel ratio to the lean side at the time of catalyst deterioration diagnosis, it is possible to separate medium deterioration from large deterioration.

On the other hand, in order to forcibly shift the air-fuel ratio to the lean side during the deterioration diagnosis, the updating of the learning value LP is prohibited when the number of updates in the learning area overlapping the diagnosis area becomes a predetermined number or more. Battery backup RAM so that the air-fuel ratio shifts from the stoichiometric air-fuel ratio
Is changed to a learning value for diagnosis.
In connection with such a forced shift of the air-fuel ratio, steps 33 and 34 are added in the subroutine for reading the learning value shown in FIG. 4 and step 46 is newly added in the subroutine for updating the learning value shown in FIG. Have been.

Also, for diagnosis, when the catalyst is new,
If the air-fuel ratio has shifted to the lean side,
A slight disadvantage in terms of performance.
In step 15, it is determined whether the deterioration is large or medium.
Prior to the stoichiometric air-fuel ratio
Do it below.

FIGS. 10 and 11 are flow charts showing a catalyst deterioration diagnosis, in which the output of the upstream O ₂ sensor (or idle air) is performed only once per operation (from the start of the engine to the stop of the engine). It is executed every time the fuel ratio feedback correction coefficient α) is inverted, every certain crank angle, or every certain time.

First, steps 61, 62, and 63 are steps for checking whether or not the condition is a diagnostic condition, which is the same as the conventional case. The following conditions: (1) the diagnosis permission condition is satisfied (step 61); (2) the diagnosis area (step 62); and (3) the update count counter value LCNT in the learning area overlapping the diagnosis area is a predetermined value. It is determined whether or not LCNT0 is exceeded (step 63). If any of the conditions is not satisfied, the process proceeds to step 64, another counter value j2 is set to an initial value of 0, and if all three conditions are satisfied, It is determined that the diagnosis condition is satisfied, and the process proceeds to steps 101 and 102 .

Here, the diagnosis permission condition of <1> is that the water temperature at the start of the engine is equal to or higher than a predetermined value, that a predetermined time has elapsed since the completion of engine warm-up,
It _second sensor is active (which is determined from the downstream O ₂ sensor output level) is a case that satisfies all.

[0058] The above diagnostic region of <2>, it is the air-fuel ratio feedback control region based on the upstream side O ₂ sensor output, when the fact that the operation condition is in a steady state, satisfy both, more <A> The vehicle speed VSP is within a predetermined range, <a> the engine speed N is within a predetermined range, and <c> the basic pulse width Tp is within a predetermined range.

The update number counter value LCN of the above <3>
T is a value that is incremented by one each time the learning value LP for each learning area shown in FIG. 5 is rewritten. If the learning is not sufficiently advanced in the learning area overlapping the diagnosis area, and the average air-fuel ratio is biased to the rich side or the lean side due to output variations of the upstream O ₂ sensor, the downstream O ₂ Since the reversal cycle of the sensor output is affected and an error occurs in the reversal number ratio HZRATE, which will be described later, when LCNT ≦ LCNT0, the accuracy does not deteriorate in the deterioration diagnosis by not proceeding to step 101 and subsequent steps. That is to prevent. Therefore, as a result, of all the learning areas, the learning area in which learning has sufficiently progressed becomes the diagnosis area.

Steps 101 and 102 determine the air-fuel ratio.
Before performing the deterioration diagnosis by shifting to the
This is the part that performs a preliminary diagnosis. Taimi whose diagnostic conditions are met
Immediately without lean-shifting the air-fuel ratio
Calculate the inversion frequency ratio HZRATE, and determine this by a predetermined judgment.
It is compared with the reference value CNGHZ1 (steps 101 and 10).
2) If HZRATE <CNGHZ1, the catalyst is medium
Diagnosis is terminated because it has not progressed to deterioration, and HZR
When ATE ≧ CNGHZ1, performance degradation is more than medium degradation
Judgment that the air-fuel ratio is lean-shifted
The process proceeds to disconnection (that is, step 65 and subsequent steps).

In step 65, it is determined whether the diagnosis condition has been satisfied for the first time.
To set the flag for inhibiting the update of the learning value LP to "1".
In step 67, the change flag to the diagnostic learning value is set to "1", and after a certain period, step 6 is executed.
Return to 1. Note that both flags are initially set to “0”.

When the update prohibition flag is set to “1”, in the learning value update subroutine of FIG.
This makes it impossible to proceed thereafter, and the updating of the learning value LP is prohibited.

When the change flag to the diagnostic learning value is set to "1", the process proceeds from step 33 to step 34 in the learning value read subroutine of FIG. 4, where the learning value is LP = LP-LPCNG. (9) However, it is changed to a value smaller by a certain value in the formula of LPCNG: positive predetermined value. Since the learning value LP on the right side of the equation (9) is a value after learning has sufficiently progressed,
The control air-fuel ratio (average air-fuel ratio) based on the learning value LP on the right side of the equation (9) should be close to the stoichiometric air-fuel ratio. Accordingly, by subtracting LPCNG from the learning value LP on the right side of the equation (9) as a diagnostic learning value (that is, the learning value LP on the left side of the equation (9)), the control air-fuel ratio is calculated from the stoichiometric air-fuel ratio by LP.
The air-fuel ratio is shifted toward the lean side by a predetermined value determined by CNG.

In FIG. 10, if the diagnosis condition is not satisfied for the first time, the routine proceeds to step 68, where the counter value j2 is incremented. In step 69, the counter value j2 is compared with a predetermined value n2, and if j2 <n2, if there is,
Prepare for the next cycle.

If the diagnosis condition is maintained, the condition j2.gtoreq.n2 is reached in step 69, and the routine proceeds to step 70 in FIG. The reason for not proceeding to step 70 until j2 ≧ n2 is to wait for the control air-fuel ratio to settle to a lean value outside the stoichiometric air-fuel ratio in the forced lean shift of the air-fuel ratio. is there. FIG. 10 is executed every time the output of the upstream O ₂ sensor (or the air-fuel ratio feedback correction coefficient α) is inverted, every fixed crank angle, or every fixed time.
The case where ≧ n2 is satisfied is when the output of the upstream O ₂ sensor (or the air-fuel ratio feedback correction coefficient α) has been inverted a predetermined number of times, when a predetermined crank angle has elapsed, or when a predetermined time has elapsed.

In FIG. 11, in step 70, the inversion frequency ratio HZR of the upstream O ₂ sensor and the downstream O ₂ sensor
The ATE HZRATE = f2 / f1 ... ( 10) However, f2: the downstream _{O 2} sensor rich, lean inversion frequency f1: the upstream _{O 2} sensor rich is calculated by the formula of the lean inversion frequency. Incidentally, it is of course possible to measure the inversion cycle of each O ₂ sensor and determine the inversion frequency ratio HZRATE from these.

In step 71, the inversion frequency ratio HZRATE in equation (10) and a predetermined judgment reference value (for example, 0.4) C
NGHZ is compared, and if HZRATE <CNGHZ, it is determined that the catalyst is in a moderately degraded state, and the counter value CCATNG is reset to the initial value of 0 in step 72 without turning on the warning light, and then update is prohibited in step 73. The flag and the change flag to the diagnostic learning value are both reset to “0”, and the diagnosis is terminated.

On the other hand, at step 71, HZRATE ≧ CN
When GHZ is reached, the routine proceeds to step 74, where the counter value C
CATNG is incremented and compared with a predetermined value (for example, 3) CCATJ in step 75. If HZRATE ≧ CNGHZ in step 75, it is determined that the catalyst has advanced to the degree of great deterioration, and a warning lamp is turned on in step 76 to terminate the diagnosis.

As described above, when diagnosing catalyst deterioration using the inversion frequency ratio HZRATE of the upstream O ₂ sensor and the downstream O ₂ sensor, the control air-fuel ratio is shifted from the stoichiometric air-fuel ratio to the lean side by diagnosing the catalyst. In the vicinity of the air-fuel ratio, the degree of medium deterioration that cannot be completely separated from large deterioration can also be accurately separated, so that the catalyst can work to the maximum extent immediately before proceeding to large deterioration.

When the performance of the catalyst is deteriorating more than the degree of moderate deterioration, even if the air-fuel ratio is lean-shifted for diagnosis, the exhaust performance does not extremely decrease.

Also, the air-fuel ratio for the diagnosis is limited only when the learning is sufficiently advanced in the learning area overlapping the diagnosis area (that is, when the average air-fuel ratio is accurately maintained near the stoichiometric air-fuel ratio). When the lean shift is performed, the lean shift amount of the air-fuel ratio is stabilized, so that the separation accuracy between the medium deterioration and the large deterioration is improved.

For example, when the average air-fuel ratio originally deviates to the rich side due to the output variation of the O ₂ sensor or the like, and the learning is not sufficiently advanced in the learning area overlapping the diagnosis area, the O ₂ sensor is not used. Since the lean shift amount at the time of diagnosis is insufficient by the rich amount of the air-fuel ratio due to output variation or the like, this O ₂
If there is no rich portion of the air-fuel ratio due to variations in the output of the sensor, the value of HZRATE is set to the judgment reference value CNGHZ.
Although it was smaller (that is, diagnosed as medium deterioration), HZRATE
Is not smaller than CNGHZ (that is, it is diagnosed as large deterioration). On the other hand, after the learning has sufficiently progressed in the learning area overlapping the diagnosis area, the learning value LP eliminates the rich portion of the air-fuel ratio due to the variation in the output of the O ₂ sensor, etc. The amount is not short, and therefore, it is possible to avoid erroneous diagnosis that the catalyst is moderately deteriorated but has advanced to a large degree.

Further , a new catalyst is used for the purpose of deterioration diagnosis.
Sometimes the air-fuel ratio is shifted to lean or rich.
Would be a little disadvantageous in terms of exhaust performance
However, according to this example, the catalyst deteriorates more than the medium deterioration.
Until the cause, shift the air-fuel ratio to rie down side and the rich side
The air-fuel ratio does not deviate from the stoichiometric air-fuel ratio when new
Can be avoided.

FIG. 12 shows a second embodiment.
Corresponds to 1.

In this example, when HZRATE ≧ CNGHZ, the process proceeds to step 121, and the weighted average value HZRTAV of the inversion frequency ratio HZRATE is calculated as follows: HZRTAV = HZRTAV- ₁ × (1-K2) + HZRATE × K2 (13) where K2: weight Average coefficient (0 ≦ K2 <1) HZRTAV ₋₁ : Calculated by the formula of the previous weighted average value, and the weighted average value HZRTAV is compared with the determination reference value CNGST in step 122, and HZRTAV is calculated.
If <CNGST, the weighted average is repeated and HZRT
When AV ≧ CNGST, the routine proceeds to step 76, where the warning light is turned on.

FIG. 13 is a subroutine for obtaining the above-mentioned judgment reference value CNGST, which is executed every time the inversion frequency ratio HZRATE is calculated. The weighted average counts up the number of times that at step 131, from the weighted average number in step 132 with reference to the table of Figure 14 the content obtaining the criterion value CNGST.

The criterion value CNGST is gradually decreased as the weighted average number increases. This is because the weighted average value HZRTAV of the inversion number ratio increases as the weighted average number increases.
Is increased, so that the reference value (that is, CNGST) for determining large deterioration can be reduced.

In this example, since the weighted average value of the number of times of reversal is used for the deterioration diagnosis, the accuracy of the diagnosis is improved as compared with the first embodiment.

[0079] In the embodiment, although shifting the air-fuel ratio to the lean side or the rich side by the change of the two proportional amount P _L and P _R, by changing the proportional portion P _L or P _R only one, also integral portion changes in both or only one of I _L and I _R may be shifted to the air-fuel ratio to the lean side or the rich side.

Further, the air-fuel ratio can be shifted to a lean side or a rich side by other means, such as by directly reducing or increasing the amount of fuel or air. For example, by opening the auxiliary air valve provided in the passage bypassing the throttle valve at a predetermined opening, the air-fuel ratio can be shifted toward the lean side by a predetermined amount.

[0081]

According to the first aspect of the present invention, there is provided a means for calculating a basic injection amount in accordance with the operating conditions of an engine, a sensor for outputting an output in accordance with the oxygen concentration in exhaust gas upstream of the catalyst, and Means for calculating a feedback constant based on the output, means for storing a learning value for the feedback constant, means for correcting the feedback constant with a learning value read from the storage means, and using the corrected feedback constant. Means for calculating the air-fuel ratio feedback correction amount, means for calculating the fuel injection amount by correcting the basic injection amount with the air-fuel ratio feedback correction amount, means for supplying the fuel of this injection amount to the intake pipe, A sensor that outputs according to the oxygen concentration in the exhaust gas on the downstream side of the catalyst, and the stoichiometric air-fuel ratio based on the output of the downstream sensor. Means for updating the learning value of the storage means in the direction of returning to the side, means for determining whether or not the number of updates by the updating means is a predetermined number or more. Means for comparing the two sensor outputs to determine whether the catalyst has undergone a small deterioration or a medium deterioration or more, and based on the result of the determination, when the catalyst has a medium deterioration or more, the updating of the learning value by the updating means. Means for prohibiting and changing the learning value read from the storage means to a learning value for diagnosis so that the air-fuel ratio deviates from the stoichiometric air-fuel ratio; and an air-fuel ratio state deviated from the stoichiometric air-fuel ratio by the learning value for diagnosis. Means for comparing the outputs of the two sensors to determine whether the catalyst has undergone major or moderate deterioration.
It is possible to work to the fullest extent before,
Check if the performance degradation of the catalyst is progress to more degree of deterioration
Even if the fuel ratio is deviated from the stoichiometric air-fuel ratio, the exhaust performance deteriorates extremely
And the number of updates has exceeded the specified number
In the timing, the output variation of the upstream sensor and the catalyst
Air-fuel ratio variation due to non-equilibrium gas on the upstream side
Regardless of these, the air-fuel ratio is centered around the stoichiometric air-fuel ratio.
The cycle is changed by
In this case, the amount deviated from the stoichiometric air-fuel ratio is stabilized,
Separation accuracy of deterioration and large deterioration is improved, and new catalyst
Sometimes the air-fuel ratio does not deviate from the stoichiometric air-fuel ratio
From, do not be a disadvantage in terms of the exhaust performance.

The second invention comprises means for calculating the basic injection amount according to the operating conditions of the engine, a sensor for outputting an oxygen concentration in the exhaust gas upstream of the catalyst, and an output of the upstream sensor. Means for calculating a feedback constant based on the above, means for storing a learning value for the feedback constant independently for each learning area divided into a plurality of areas using at least the engine speed and load as parameters, and an engine at that time. Means for reading a learning value of a learning area to which the rotational speed and load belong, means for correcting the feedback constant with the read learning value, and means for calculating an air-fuel ratio feedback correction amount using the corrected feedback constant And calculating the fuel injection amount by correcting the basic injection amount with the air-fuel ratio feedback correction amount. Means for supplying the fuel of this injection amount to the intake pipe, a sensor for outputting an output corresponding to the oxygen concentration in the exhaust gas downstream of the catalyst, and the engine speed and load at that time are stored in the same learning area in the same learning area. Means for updating the learning value of the same learning area during the period to return the air-fuel ratio to the stoichiometric air-fuel ratio side based on the output of the downstream sensor, and a learning area overlapping the diagnostic area by the updating means. Means for judging whether or not the number of updates is equal to or more than a predetermined number. When the number of updates is equal to or more than a predetermined number from the determination result, the outputs of the two sensors are compared to determine that the catalyst has a small deterioration in the learning area overlapping the diagnosis area. Means for determining which of the medium deterioration or more has occurred, and a diagnosis area by the updating means when the catalyst has a medium deterioration or more in a learning area overlapping the diagnosis area from the determination result. Means for prohibiting the update of the learning value in the learning area, and changing the learning value read from the learning area overlapping the diagnosis area by the storage means to the learning value for diagnosis so that the air-fuel ratio deviates from the stoichiometric air-fuel ratio. Means for comparing the two sensor outputs in an air-fuel ratio state deviating from the stoichiometric air-fuel ratio based on the learning value for diagnosis to determine whether the catalyst has undergone large deterioration or medium deterioration . In addition to the effects of the first invention,
Medium degradation due to higher accuracy of the learning value than the first invention
And the separation accuracy of large deterioration is improved .

[0083] The third invention is Te placed <br/> the first or second inventions, the one of the determining means is one of the major degradation and medium deterioration occurs, the number of reversals or inversion periods from the respective sensor output A means for calculating, a means for calculating the ratio or difference of the number of inversions or the inversion cycle, a means for calculating a weighted average of the ratio or the difference, and a means for comparing the weighted average with the determination reference value. since, in addition to the effects of the first or second aspect of the invention, it improves the middle degradation and separation accuracy of the large degradation further.

[Brief description of the drawings]

FIG. 1 is a system diagram of an embodiment.

FIG. 2 is a flowchart for explaining calculation of a fuel injection pulse width Ti.

FIG. 3 is a flowchart for explaining calculation of an air-fuel ratio feedback correction coefficient α.

FIG. 4 is a flowchart for explaining reading of a learning value LP;

FIG. 5 is a diagram illustrating a learning area.

FIG. 6 is a flowchart for explaining updating of a learning value LP.

FIG. 7 is a graph showing characteristics of oxygen concentration, CO concentration, and O ₂ sensor output downstream of a catalyst, in contrast to a case of medium deterioration and a case of large deterioration.

FIG. 8 is a waveform diagram showing a case where the air-fuel ratio is lean-shifted, showing a case where the deterioration is moderate and a case where the deterioration is large.

FIG. 9 is a characteristic diagram of an inversion frequency ratio HZRATE with respect to an air-fuel ratio.

FIG. 10 is a flowchart for explaining deterioration diagnosis.

FIG. 11 is a flowchart for explaining deterioration diagnosis.

FIG. 12 is a flowchart for explaining deterioration diagnosis of the second embodiment.

FIG. 13 is a flowchart for explaining calculation of a determination reference value CNGST in the second embodiment.

FIG. 14 is a characteristic diagram showing the contents of a table of determination reference values CNGST.

FIG. 15 is a waveform diagram of a conventional O ₂ sensor.

FIG. 16 is a diagram corresponding to the claims of the first invention.

FIG. 17 is a diagram corresponding to a claim of the second invention.

FIG. 18 is a diagram corresponding to claims of the third invention.

[Explanation of symbols]

Reference Signs List 11 air flow meter 12 crank angle sensor 13 O ₂ sensor (upstream sensor) 14 O ₂ sensor (downstream sensor) 15 control unit 21 basic injection amount calculation means 22 upstream sensor 23 feedback constant calculation means 24 air-fuel ratio feedback correction amount Calculation means 25 Fuel injection amount calculation means 26 Fuel supply means 27 Downstream sensor 29 Large / medium deterioration determination means 31 Medium deterioration or more determination means 41 Learning value storage means 42 Feedback constant correction means 43 Learning value update means 44 Update count determination means 45 Update prohibition / learning value changing means 51 Learning value storage means 52 Learning value reading means 53 Learning value updating means 54 Update count judging means 55 Update prohibition / learning value changing means 61, 62 Inversion count / cycle calculation means 63 Inversion count / cycle ratio / Difference calculation means 6 The weighted average value calculating means 65 comparison means

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI F02D 45/00 340 F02D 45/00 340F G05B 23/02 302 G05B 23/02 302V

Claims (3)

(57) [Claims] 1. A means for calculating a basic injection amount according to operating conditions of an engine; a sensor for outputting an oxygen concentration in an exhaust gas upstream of a catalyst; and a feedback constant based on an output of the upstream sensor. Means for calculating a feedback constant, means for storing a learning value for the feedback constant, means for correcting the feedback constant with a learning value read from the storage means, and an air-fuel ratio feedback correction amount using the corrected feedback constant. Means for calculating a fuel injection amount by correcting the basic injection amount with the air-fuel ratio feedback correction amount; means for supplying fuel of the injection amount to the intake pipe; and exhaust gas on the downstream side of the catalyst. A sensor that outputs according to the oxygen concentration in the air, and returns the air-fuel ratio to the stoichiometric air-fuel ratio based on the output of this downstream sensor. Means for updating the learning value of the storage means in a direction, means for determining whether or not the number of updates by the updating means is equal to or greater than a predetermined number, and when the number of updates is greater than or equal to the predetermined number, the two sensor outputs Means for comparing whether small deterioration or medium deterioration or more has occurred in the catalyst, and when the medium deterioration or more has occurred from the determination result, updating of the learning value by the updating means is prohibited, Means for changing a learning value read from the storage means to a learning value for diagnosis so as to deviate the air-fuel ratio from the stoichiometric air-fuel ratio; and the two values in the air-fuel ratio state deviated from the stoichiometric air-fuel ratio by the learning value for diagnosis. Means for comparing the sensor outputs to determine whether the catalyst has deteriorated significantly or has deteriorated. 2. A means for calculating a basic injection amount according to operating conditions of an engine; a sensor for outputting an oxygen concentration in exhaust gas upstream of a catalyst; and a feedback constant based on an output of the upstream sensor. Means for calculating a learning value for the feedback constant independently for each learning area divided into a plurality of areas using at least the engine speed and load as parameters, and the engine speed and load at that time. Means for reading a learning value of a learning area to which the following belongs; means for correcting the feedback constant with the read learning value; means for calculating an air-fuel ratio feedback correction amount using the corrected feedback constant; Means for correcting the basic injection amount with a fuel ratio feedback correction amount to calculate a fuel injection amount; Means for supplying an injection amount of fuel to the intake pipe; a sensor for outputting an output according to the oxygen concentration in the exhaust gas downstream of the catalyst; and the engine speed and load at that time staying in the same learning area for a predetermined period. Means for updating the learning value of the same learning area in the direction in which the air-fuel ratio returns to the stoichiometric air-fuel ratio based on the output of the downstream sensor, and updating the learning area overlapping the diagnostic area by this updating means. Means for determining whether or not the number of times is equal to or more than a predetermined number of times. When the number of times of update is equal to or more than the predetermined number of times, the two sensor outputs are compared, and the catalyst has a small deterioration and a medium or more deterioration in the learning area overlapping the diagnosis area. Means for determining which of the above has occurred, and when the catalyst has a medium deterioration or more in the learning area overlapping with the diagnosis area based on the determination result, it overlaps with the diagnosis area by the updating means. Means for prohibiting the updating of the learning value in the learning area, and changing the learning value read from the learning area overlapping the diagnosis area by the storage means to a learning value for diagnosis so that the air-fuel ratio deviates from the stoichiometric air-fuel ratio; Means for comparing the two sensor outputs in an air-fuel ratio state deviating from the stoichiometric air-fuel ratio based on the learning value for diagnosis to determine whether the catalyst has undergone large deterioration or medium deterioration. Engine degradation diagnostic device for engine. 3. The means for determining whether the large deterioration or the medium deterioration has occurred is means for calculating the number of inversions or the inversion cycle from the output of each sensor, and means for calculating the ratio or difference between the number of inversions or the inversion cycle. If, means for calculating a weighted average value of this ratio or difference, claim 1, characterized in that it consists of a means for comparing the criterion value the weighted average value or
3. The apparatus for diagnosing catalyst deterioration of an engine according to claim 2 .