JPH0219634A - Control unit for internal combustion engine - Google Patents

Abstract

PURPOSE:To make an air-fuel ratio accurately controllable to such one at the lean side securing the maximum torque by compensating and learning a desired air-fuel ratio so as to cause cylinder internal pressure to take the maximum value according to the cylinder internal pressure and exhaust temperature at the time of high load stationariness. CONSTITUTION:A control unit 35 judges a high load stationary state of an engine 21 from fuel's fundamental pulse width and engine speed. Then, a desired air-fuel ratio is compensated so as to cause cylinder internal pressure to take the maximum value according to the cylinder internal pressure being detected by each of cylinder internal pressure sensors 23A, 23B on the condition that exhaust temperature being detected by exhaust temperature sensors 33, 34 is within a time constant value at the time of high load stationariness and this exhaust temperature. In addition, this correction value is learned, while during this compensative learning, ignition timing is fixed to the specified base ignition timing. With this constitution, an air-fuel ratio can be turned to leanness as securing the maximum torque and while setting the exhaust temperature to the specified range. Accordingly, it is accurately controllable to the air-fuel ratio that secures the maximum torque.

Description

[Detailed description of the invention] (Industrial application field) The present invention is directed to an improvement in a control device for an internal combustion engine.

(Conventional technology) 8! Some control devices are equipped with a sensor that detects knocking, and when knocking occurs, the ignition timing is retarded or the air-fuel ratio is changed to the side where knocking does not occur (rich side). Regarding ignition timing control, see Japanese Patent Laid-Open No. 60-26172). Here, a level that causes a slight knocking state is set as a predetermined level, and the ignition timing and air-fuel ratio are controlled using this predetermined level as the n target value (such control is hereinafter referred to as "knock control").

) is carried out. This is because, although strong knocking that adversely affects the engine must be avoided, light knocking itself does not adversely affect the engine, and on the contrary improves fuel efficiency by improving combustion efficiency.

On the other hand, from the standpoint of improving output, it is desirable to control the ignition timing so that the maximum torque can be obtained for the same amount of fuel. Therefore, control fi1 (hereinafter referred to as "M
It is called "B T lil control". ).
(See Japanese Patent Application Laid-Open No. 16268/1983). For example, the crank angle at which the cylinder pressure of the engine is maximum (hereinafter simply referred to as the [maximum cylinder pressure crank angle]) θpmax
Alternatively, the ignition timing is corrected and controlled using a feedback amount based on the deviation between the detected value and the target value so that the detected value of the indicated mean effective pressure Pi matches a preset target value.

Further, non-king control and M+3T control are often combined, and in this case, knock control is given priority in the relationship between the two. In the following, the control of the combination of the two will be referred to as "knock-MB'r control".

Furthermore, with the aim of eliminating errors due to individual differences and changes over time, ignition timing control and air-fuel ratio control equipped with a learning function have been proposed (Japanese Patent Application Laid-Open No. 55-11111
-96339).

(Problem to be solved by the invention) By the way, when the engine is under high load, this is the operating range where knocking is likely to occur and a large output is required. Therefore, while performing knock control, the mixture is enriched to increase the output. However, even in the high load range, in steady state, no output is required in transient conditions, so if the air-fuel ratio can be made as lean as possible without degrading driving performance, unnecessary fuel consumption can be avoided. can be suppressed.

If you want to perform this kind of control here, 7tsuku-M
BT! It is necessary to control both II all and air-fuel ratio control, that is, both the ignition timing and the air-fuel ratio according to demand. However, in the conventional example, the content of knock control is to control only one of the ignition timing or the air-fuel ratio, so it is impossible to control both of them at the same time. This is because the main focus is only on avoiding 7-king, and no consideration is given to improving fuel efficiency by making the air-fuel ratio as lean as possible during steady high-load conditions.

On the other hand, when the load is steady at high load, it is necessary to take measures against heat load other than knocking, that is, to control the exhaust temperature so that it does not exceed a predetermined value. For example, if the temperature downstream of the exhaust valve exceeds a predetermined value at high speeds, or if the temperature downstream of the catalyst exceeds a predetermined value during low speeds, the engine body and catalyst are likely to be damaged.

Furthermore, it should be taken into consideration that differences in the compression ratio, the detection accuracy of the air flow sensor, or the cooling efficiency of the combustion chamber will cause variations in the air-fuel ratio and α-ignition timing, resulting in deviations from the maximum torque point and minimum fuel efficiency point. Then, it is also necessary to add a learning function.

This invention aims to solve these conventional problems.

(Means for Solving the Problem) As shown in FIG. 1, the present invention includes a means 1 for detecting cylinder pressure, a means 99.2 for detecting knocking from the cylinder pressure, and a MINT correlation quantity ( means 3 for detecting the indicated mean effective pressure or cylinder pressure maximum crank angle; means 4 for correcting the ignition timing in accordance with the detected knocking value and the detected value of the MBT correlation amount; means 5 for calculating the target air-fuel ratio, and means 6 for calculating the target air-fuel ratio;
An internal combustion engine comprising means 7 for detecting an actual air-fuel ratio, means 8 for correcting the fuel supply amount according to the difference between the actual air-fuel ratio and the l''I air-fuel ratio, and means 9 for learning this correction amount. In the control device, a hand Pii O that determines a high load steady state;
Hand fill for detecting exhaust temperature and means 1 for correcting the target air-fuel ratio so that the in-cylinder pressure takes the maximum value according to the in-cylinder pressure and exhaust temperature under the condition that the exhaust temperature is within a predetermined value during high load steady state
2, means 13 for learning this correction amount, and means 1 for fixing the ignition timing to a predetermined base ignition timing during this correction learning.
4 will be provided.

(Function) Therefore, by correcting and learning the target air-fuel ratio so that the in-cylinder pressure takes the maximum value according to the in-cylinder pressure and exhaust temperature during high load steady state, maximum torque can be ensured and the exhaust temperature can be kept within a predetermined range. By making the air-fuel ratio leaner while reducing fuel consumption, fuel consumption under high loads can be reduced.

If the ignition timing is corrected during target air-fuel ratio correction learning, the cylinder pressure will clog and the maximum torque point will change, but by fixing the ignition timing to the base ignition timing during this correction learning, , the air-fuel ratio can be precisely controlled to obtain the maximum torque, and knocking does not occur.

It goes without saying that controllability is improved by adding a learning function.

(Embodiment) An embodiment in which the present invention is applied to a V-type six-cylinder engine is shown in FIG. 2. Reference numerals 23A and 23B are washer mold internal pressure sensors that are attached to the washer of the spark plug of the engine 21. 23
The amount of charge output by B is converted into a voltage value by Charno amplifiers 24A and 24B, and the output of this charge amplifier is input to multiplexer 40 as an in-cylinder pressure signal. Using this cylinder pressure signal, the indicated average effective pressure Pi or the detection value of the cylinder pressure maximum crank angle sensor ax is calculated.

On the other hand, the charge amplifier output is from the knocking signal detection circuit 2.
5A and 25B, only the high frequency signal specific to knocking is separated, and this knocking signal 48 is also input to multiplexer 40.

29 is 43 per unit angle (for example, 1°) of the crank angle
This is a crank angle sensor that generates a signal (reference signal) at every predetermined angle (120°).
is input. The engine speed N is determined from the unit angle signal, and the cylinder is determined from the reference position signal.

31 is a sensor for detecting intake air and lQa (for example, a flap type or hot wire type air 7a-meter); 32
is a sensor that can detect not only the stoichiometric air-fuel ratio but also other air-fuel ratios, 33 and 34 are sensors that detect the exhaust temperature downstream of the exhaust valve and downstream of the catalyst, respectively, and the signals from each sensor are sent to a multiplexer 40. is input.

35 is a control unit into which various signals from sensors are input, and CPU 36. ROM37, RAM3
8 and l1039 are multi-break? 40. A/L'
) A converter 41 and a digital boat 42 are provided. The control unit 35 determines the fuel injection pulse width 'ri to be outputted and the ignition timing 5ETAD to be outputted based on the operating condition signal, the air-fuel ratio, and the ignition timing feedback signals.
V and is converted into an injection pulse and an ignition pulse, which are output to a fuel injection valve and an ignition device (not shown).

Next, the control contents executed by the control unit 35 will be explained based on FIGS. 3(A) and 3(B).
The fuel injection pulse width 'i is calculated using the following equation (1).

Ti=TpXTrbaXKLcdXLiad+Ts=l
l) However, 'l''p: Fundamental pulse width (K cto+
aL
; Air-fuel ratio learning coefficient (learning value used for normal control) Lad; Feedback coefficient Ts based on the signal from the air-fuel ratio sensor 32; The invalid pulse width of the battery correction amount and the target air-fuel ratio TflJa are calculated by the following equation (2) be done.

'r'fba=M mouth+a+D fbas+(K tw
+K as+ -)...(2) However, Mfba: base air-fuel ratio (set using N and Qa as parameters) L') rbus: air-fuel ratio sensor 1 (air-fuel ratio correction learning value D
H+ulor air fuel ratio correction fiDfba++)Ktm+K
as+...: Correction amount for water temperature increase, etc., air-fuel ratio correction amount D
fbap = air-fuel ratio correction learning value DIl + a1 - based on the signal from the exhaust temperature sensor 33.34 (air-fuel ratio correction value Dfl
+a.

Further, the ignition timing 5ETADV is calculated using the following equation (3).

S E T A D V = TadvlIl+ 5e
Lcs+ Advld - (3) However, T advn
: Base ignition timing (set using N and Qa as parameters) Setc3;+! /Fire timing correction ii (M13T correction 1i
Mbtcs or knock correction 11Advcs (small crawl) A dvld;, α fire timing learning value PjSs In the diagrams (A) and (B), in step 101, high load loading is determined based on the basic fuel pulse width 1゛p and the rotation speed N. It is determined whether t+. In step 102, it is determined whether the basic pulse width Tp and the rate of change ΔTρ and ΔN of the rotational speed N are within predetermined values and the engine cooling water temperature is a predetermined value or higher.
It is determined whether the conditions for performing learning control are satisfied.

In step 103, it is determined t'q whether learning of L B T is completed.

If the state is other than a high load steady state or if the learning of L B 'r has been completed, the process proceeds to step 123 and subsequent steps and the knock M B
Ti control (ignition timing control) and normal air-fuel ratio control (
However, if LBT learning has not been completed in a high load steady state, it is determined that the LBT learning control condition is met and the process proceeds to step 104 and subsequent steps.

In step 104, the ignition timing 5ETADV is set and controlled to the base ignition timing 'advm. That is, LB
Knock-MBT control is canceled during T learning control.

In steps 105 and 106, this operating region (Tp+N)
If the air-fuel ratio correction learned value 1) fl+al is learned for the first time, an initial value (a predetermined rich amount) is set in Dfbul. In steps 107 and 108, the operating region (1°p
*Dfl+al is read from the learning grid corresponding to N) and set to the air-fuel ratio correction coefficient D fbas of the target air-fuel ratio Tba. Then, according to the target air-fuel ratio Tfba calculated based on I) fbas. The air-fuel ratio (fuel injection jl T i ) is controlled in step 128, and when the predetermined period A (step 107) set by the timer has elapsed, the air-fuel ratio (fuel injection jl T i ) is 10
Proceed to 9 or below. Period A corresponds to the time until the outputs of the exhaust gas temperature sensors 33 and 34 become stable after the air-fuel ratio is changed.

Steps 109 and 110 include the difference between the exhaust limit temperature LexLe pass downstream of the exhaust valve and the exhaust temperature 1xLemp, and the difference between the exhaust limit temperature Lcatteup and the exhaust temperature C
The difference in utteB is calculated, and depending on the smaller difference, the air-fuel ratio? 1Ilf value DfL+a is calculated. In step 111, the air-fuel ratio correction coefficient Dfbas set in step 108 is compared with the added value of the base air-fuel ratio Mfba and Dfba, and if they are not equal, the process proceeds to step 114 from the air-fuel ratio correction learning value Dfba to I). fb
The air-fuel ratio correction 1iDfbap obtained by subtracting a is set to the air-fuel ratio correction coefficient DI'bas. Then, the air-fuel ratio is controlled in step 128 according to the target air-fuel ratio Tfba calculated based on Dfbaa, and the timer controls the air-fuel ratio for a predetermined period B.
After (step 109) has passed, the process proceeds to step 115 and subsequent steps.

In the case of rabbits, the air-fuel ratio in period B is made leaner by the air-fuel ratio correction value D"ba based on the exhaust temperature than the air-fuel ratio in period A. Note that in step 111, the added value coincides with the lean limit value X. If so, step 112.113
At this point, DflJa is set to 0, and the LBT learning control is ended at that point.

In steps 115 and 116, the indicated mean effective pressure Pi is calculated from the cylinder pressure using the following equation (4). This calculation is performed for a predetermined period C by a timer, and the average value during this period is determined as Pi.

P i = 1 / V engineering, 'pvdθ However, ■: Schilling volume between predetermined crank angles p; Cylinder pressure at a predetermined crank angle V: Schilling volume at a predetermined crank angle In this case, ignition timing S E T Since ADV is set as the base ignition timing T advta (step 1
04), the error of Pi becomes small.

In step 117, this indicated mean effective pressure Pi is compared with the previous learned value Pi-1, and if Pi≧P1-1, in step 118, the air-fuel ratio correction amount Dfbap obtained in step 114 is adjusted to the operating range ('rp , N) in the memory as the air-fuel ratio correction learning value Dfbal, that is, D
fbul is updated, and this Dfbul is set to the air-fuel ratio correction coefficient D_fbus. In step 119, the previous learned value Pi-H is updated to the current Pi.

Then, the target air-fuel ratio 'r calculated based on DI'bas
The air-fuel ratio is controlled in step 128 according to fba, and when a predetermined period C has elapsed, step 122
The timer is cleared and steps 101 to 119 are performed again.
The control up to this point is repeated.

That is, within a range where the exhaust temperature and air-fuel ratio do not exceed the limit temperature and lean limit, the air-fuel ratio is made lean according to the exhaust temperature and the indicated mean effective pressure Pi, and at the same time, the lean state is made so that Pi becomes maximum, that is, the maximum torque is achieved. Search the air-fuel ratio on the side.

Then, if pi<pi-1 in step 117,
The air-fuel ratio just before Pi<Pi-l becomes the lean air-fuel ratio (LBT) that provides the maximum torque, and at this time, the air-fuel ratio correction learning value Dfbal is not updated, and in step 120 the previous Dfba17>'D fbas is reset, and in step 121, LBTBT learning control is completed.

On the other hand, in a state other than a high load steady state or after the L B T learning control ends, the process proceeds to step 123 and subsequent steps, and the knock = MB
T control is performed.

When non-king does not occur, the MBT correction amount Mbtcs is calculated according to the deviation between the cylinder pressure maximum crank angle θl]II a The ignition timing correction amount S eLcs is set to advance or retard the ignition timing. When knocking occurs, a knock correction 11Advcs is calculated according to the detected value of the knocking, and the knocking correction 11Advcs is set as the ignition timing correction amount Setcs to retard the ignition timing 71''.

Then, in the case of MBT control, step 124.12
In step 5, the ignition timing learning value Advld is updated by the following equation (5) using the correction amount M bias.

Advld=Advld-H+(Ml+Lcs-Ad
vlcl -1)X G min -(5) However, A dvld -1* +1fT times learning value G
a1n Nigain Also, in the case of knock control, steps 124 to 126
Knock allowance MAAD set according to the rotation speed N at
V is read and the ignition timing learning value Adv! d is the knock correction and 1iAdvcs is updated to the following equation (6).

Advld= Advld −+ 10(Advcs −
AdvldM A A D V )X Gain
-(6) Note that in steps 128 and below, the air-fuel ratio is flNI-controlled according to the feedback coefficient 1-md of the air-fuel ratio sensor 32 so that the target air-fuel ratio TI'ba is reached, and the air-fuel ratio learning coefficient KLcd is used. Normal air-fuel ratio learning control is performed. In this case, if ΔT1)1ΔN is within a predetermined value, the fi cooling water temperature is above a predetermined value, and the feedback coefficient L+ad is reversed at least once, L
The air-fuel ratio learning coefficient KLcd is calculated according to the deviation of +ad by the following formula (
7) will be updated.

K Lcd= K Lcd −1+(Lead
-1,0) I will explain based on this.

When a predetermined high load steady state is reached, LBTBT learning control is started.1. The flaming fire timing is fixed to the base flaming timing, and the air-fuel ratio is set to an initial value. For example, if this control point is a
Point.

Then, while comparing the indicated mean effective pressure Pi from point a,
In addition, the air-fuel ratio is made lean while taking into account the exhaust temperature limit and the lean limit, and is learned, and when Pi becomes the maximum at point C by comparison with point b, the air-fuel ratio is controlled to the air-fuel ratio at point C. The LBTBT learning control is completed.

Next 1, [, When the BT learning control is finished, 7 nok-M
INT control is activated and 5α fire 11.2 period is advanced from point C. In the case of point C, knock control is applied and the ignition timing is controlled to dA'C, which is the 7 knock limit, and the knock margin (Moe (6) formula) is calculated from the control amount (2,
-1"i is the learning value of the ignition timing. Also, when advancing the ignition timing to move toward the M 13 T line, the M
Ignition timing control and learning is performed using B'I'' control.

In this way, LBT learning is performed according to the high load steady state at' indicated mean effective pressure and exhaust temperature, and this ':
#- During training, the ignition timing is fixed at the base ignition timing, so
The indicated mean effective pressure does not fluctuate due to the knock-M B T *lI II, so that the lean air-fuel ratio that provides the maximum torque can be accurately searched and learned within an appropriate exhaust temperature range. And L B T
After learning, the ignition timing is learned by knock-MBT control.

Therefore, maximum torque can be ensured under high loads, and the air-fuel ratio can be made lean without exhaust temperature exceeding the limit, thereby sufficiently improving fuel efficiency and ensuring optimal driving conditions. I can do it.

(Effects of the Invention) As described above, according to the present invention, there are provided a means for detecting cylinder pressure, a means for detecting knocking from the cylinder pressure, a means for detecting the MBT correlation amount from the cylinder pressure, and a means for detecting the knocking detection value and the MBT correlation amount. It comprises a means for correcting ignition timing according to a detected value of the correlation amount, a means for learning this correction amount, a means for calculating a target air-fuel ratio, a means for detecting an actual air-fuel ratio, and a means for calculating an actual air-fuel ratio. A control device for an internal combustion engine, comprising means for correcting a fuel supply amount according to a difference in target air-fuel ratio, and means for learning the correction amount, including means for determining a high-load steady state for 10 times, and detecting exhaust temperature. means for correcting the target air-fuel ratio so that the in-cylinder pressure takes a maximum value according to the in-cylinder pressure and the exhaust temperature under the condition that the exhaust temperature is within a predetermined value during steady state high load; and means for learning this correction amount. We have also provided a means to fix the ignition timing to a predetermined base value during this correction learning, so that the air-fuel ratio can be accurately adjusted to the lean side to obtain maximum torque while maintaining the exhaust temperature in an appropriate state during high loads. It can be controlled well and fuel efficiency can be significantly improved while maintaining good driving performance.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the present invention, Fig. 2 is a block diagram showing an embodiment of the present invention, and Fig. 3 (A)a (B) shows control contents.
0-chart, FIGS. 4(A) and 4(B) are explanatory diagrams showing the control state. 23A, 23B... Cylinder pressure sensor, 29... Crank angle sensor, 31... Intake air amount sensor, 32...
Air-fuel ratio sensor, 33.34... Exhaust temperature sensor, 35
···control unit.

Claims (1)

[Claims] 1. Means for detecting in-cylinder pressure, means for detecting knocking from in-cylinder pressure, means for detecting MBT correlation amount from in-cylinder pressure, and correcting ignition timing according to the detected value of knocking and the detected value of MBT correlation amount. and a means for learning the correction amount, a means for calculating the target air-fuel ratio, a means for detecting the actual air-fuel ratio, and correcting the fuel supply amount according to the difference between the actual air-fuel ratio and the target air-fuel ratio. A control device for an internal combustion engine comprising a means for determining a steady state of high load, a means for determining a steady state of high load, a means of detecting exhaust temperature, and a means of determining whether the exhaust temperature is within a predetermined value during steady state of high load. Means for correcting the target air-fuel ratio so that the in-cylinder pressure takes the maximum value according to the in-cylinder pressure and exhaust temperature under the conditions of 1. A control device for an internal combustion engine, comprising means for fixing the timing.