JPH01315647A - Knocking controller of internal combustion engine - Google Patents

Abstract

PURPOSE:To improve the stability and the responsiveness of the average of the knocking intensity values by providing an update quantity setting means which sets the update quantity per one averaging operation of an averaging means for the knocking intensity values to a greater value for the specified period after the engine state satisfies the specific requirements. CONSTITUTION:A detecting means B detects knocking intensity values from the outputs of a knocking sensor A and the knocking intensity values are averaged by an averaging means C. Based on the average of the knocking intensity values, a knocking decision level generating means D generates a knocking decision level and based on said decision level, a knocking decision means E decides the occurrence of a knocking. And according to the result from the decision, a knocking control means F controls a knocking control factor. In this case a decision means H decides whether or not the specified time passed after the engine state detected by an engine state detecting means G satisfies the specific requirements and if the satisfaction is made within the specified period, the update quantity per one averaging operation of the averaging means C is set higher by an update quantity setting means I.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention detects knock occurring in an internal combustion engine and controls knock control factors such as ignition timing and A/F (air-fuel ratio). Regarding the system.

[Conventional technology]

-Knock judgment level■ for boat fishing. is the knock sensor signal■
It is created by multiplying the averaged value v1°, 7 by a constant K (for example, Japanese Patent Laid-Open No. 180766/1983).

Further, a device has been invented that corrects VKS based on the cumulative percentage point vP of the knock sensor signal and the amount A corresponding to the variance of the distribution (for example, Japanese Patent Application Laid-Open No. 60-243369).

[Problem to be solved by the invention]

In these items, ■1゜1. , or ■, responsiveness and stability are very important. For example, if you design with an emphasis on stability, sensor signals during racing etc.
■, under rapidly changing conditions. s becomes smaller than the required value, and even though there is no knock, it is determined that there is a knock and the engine retards the angle control. On the other hand, if we emphasize responsiveness,
Even in a steady state, the variation in ■□, 0 becomes large, so the error from the true average value becomes large (and the noise may be judged as a knock, or conversely, even though knocking has occurred). vinegar .

The frequency with which knock detection becomes impossible increases. This situation also applies to (2) used to correct VKS.

As described above, the current method of creating average values or cumulative percentage points has the problem that if emphasis is placed on stability, responsiveness will be impaired, and if emphasis is placed on responsiveness, stability will be impaired.

In order to solve the above problem, the present inventors have obtained an important fact by theoretically analyzing the responsiveness of (1)61. This will be explained below.

First, assume an accelerated driving mode as shown in FIG. 2(a). According to the experimental results of the present invention etc., the knock sensor signal V
is V-V. N- Old... (1)
It can be expressed as Here, Vo and β are constants, and βzaki2.0
It is. If the number of ignitions in one cylinder is n, then dn dt dn Ne・
...(2) becomes. However, N is the unit of rpm. That is, dN and /dn have a characteristic that the larger N becomes, the smaller they become, and change as shown in FIG. 2(b).

Approximating with β = 2.0, d V / d n becomes dn
dN, dn (3). That is, V changes linearly with n, and the second
It will look like the broken line in Figure (C).

For this required value ■, (where N is an integer such as 4.8.16), ■,
. 7 changes as shown by the solid line in FIG. 2(C). This N
By V. 7 stability and responsiveness can be adjusted.

On the other hand, FIG. 3 assumes a slow acceleration mode in which dN, /dt=α is small. If N is made smaller to increase the responsiveness, the stability becomes extremely poor as shown by the solid line in FIG. 3(b).

From these facts, an easy idea comes to mind: first, when α is large, N is made small to improve responsiveness, and when α is small, N is made large to improve stability.

For example, if α≧150Orpm/s, N=4, α<1
If the speed is 50 orpm/s, N=8.

However, this method does not achieve both stability and responsiveness. This is because when N=4, although the response delay is reduced, the variation in VIIean itself becomes large, so that it cannot be expected to have much of an effect in reducing the error from the required value.

Therefore, the present inventors investigated in detail the ratio γ of the calculated value to the required value during sudden acceleration, where response delay is a problem. Figure 4(a) shows α= from N,=120Orpm
This assumes an acceleration mode of 150 rpm/s. FIG. 4(b) shows this moted version. An important thing was found from this Figure 4(b). This means that T is small (response delay is large) only in the initial stage of acceleration, and even if N is large (even if n increases, V tea□0 will converge to the required value).

That is, in the latter stage of acceleration, although the required value changes in the same way as in the early stage of acceleration (because the response delay is sufficiently small compared to the variation in the steady state), the behavior is exactly the same as in the steady state. In steady state, the larger N is,
Variations in calculated values are reduced. If calculations are performed in this region while keeping N small (if N is small), the variation in the steady state will be large, and the error from the required value will also become large. Therefore, only at the initial stage of acceleration with a large response delay, N is reduced to improve responsiveness, and then V mea
In a region where the variation in n is larger than the response delay, N may be increased to improve stability. Figure 2(b)
To explain this using an example, the area where n≦32 from the start of acceleration is set to N=4, 32<n≦64 is set to N=8, and n>64 is set to N
=16. That is, reducing N even to the region where the response delay is small is not only meaningless but also impairs stability.

Let me explain this in a little more detail. The error e between the required value and the calculated value is due to the response delay V, (this is the error when the input value has no variation at all and the required value is always input, and is 0 in a steady state) and the variation in the human power value. Variation of calculation V
, is. When the required value changes more than the variation in the calculated value during steady state (Vr )Vs ), 6'=V, +
■*', Vr. For example, if N=4.8 is compared, e4<e, also from the example of FIG. 4. However, each subscript 4.8 represents that N=4.8. However, when n becomes large and the response delay becomes smaller than the variation (V,)V,).

e-V, +V,! =:V.

becomes. In this case, the larger N, the smaller vs becomes, so naturally ell<e4 (of course, e
,,<e6).

In other words, at the initial stage when the input value changes, e, < ell
However, as n becomes larger, the relationship becomes e4>e, which is reversed. In this way, even when the required value is changing, there are regions in which a smaller N is advantageous and states in which a larger N is advantageous.

If N is kept small even in this latter region, stability will be compromised.

[Means to solve the problem]

Therefore, as shown in FIG. 1, the present invention provides a knock sensor for detecting knock of an internal combustion engine, a knock intensity value detection means for detecting a knock intensity value from this knock sensor signal, and an average of the knock intensity value. at least one of averaging means for detecting the cumulative percentage points of the distribution of knock intensity values or cumulative percentage point detecting means for detecting the cumulative percentage points of the distribution of knock intensity values; a knock judgment level generating means; and a knock judgment level based on the knock judgment level. A knock control device for an internal combustion engine comprising a knock determination means and a knock control means for controlling knock control factors such as ignition timing and air-fuel ratio according to the knock determination result, comprising: an engine state detection means for detecting an engine state; , a period determination means for determining whether a predetermined period has elapsed since the engine state reached a specific condition, and an update amount of at least one of the averaging means or the cumulative percentage point detection means only during this predetermined period. The present invention provides a knock control device for an internal combustion engine, characterized in that the knock control device includes update amount setting means for increasing the update amount.

Furthermore, it may be configured to include a learning means for learning at least one of the knock determination level and the control amount of the knock control factor, and a learning prohibition means for prohibiting learning by the learning means during a predetermined period by the period determining means.

[Effect]

As a result, the period determining means determines whether a predetermined period of time has passed since the engine state reached a specific condition by the engine state detecting means, and the averaging means or the knock strength value is adjusted only during this predetermined period. The update amount setting means increases the update amount of at least one of the cumulative percentage point detection means for detecting the cumulative percentage points of the distribution.

Further, at least one of the knock determination level or the control amount of the knock control factor can be learned by the learning means, and the learning by the learning means can be prohibited by the learning prohibition means within a predetermined period determined by the period determining means.

〔·Example〕

An embodiment of the present invention will be described below. FIG. 5 is a block diagram of an apparatus for carrying out the present invention. IO is a knock sensor, 20 is a filter that extracts a knock component from the knock sensor signal, 30 and 40 are amplifiers that amplify the signal, 50 is a knock detection microcomputer that makes a knock determination,
60 is an engine control microcomputer that calculates ignition timing and fuel injection amount from the knock judgment result from 50 and information from various other sensors 8° (crank angle sensor, pressure sensor, water temperature sensor, etc.); 70 is this microcomputer 50; An igniter that realizes the result calculated by
Injectors, etc. Further, in the figure, YGI and YG4 represent A/D conversion ports of the knock detection microcomputer 50. Note that both microcomputers 50 and 60 exchange information through the boat.

FIG. 6 shows the timer interrupt routine. This interrupt occurs when the start of knock strength value detection is at a predetermined crank angle (for example, 1
0° CAATDC) according to a flowchart not shown. When the timer interrupt routine starts from step 100, first, at step TIO, A
/D boat is cent. In step T20, the knock intensity value ■ is detected. This knock intensity value ■ is, for example, the maximum value V of the knock sensor signal as shown in FIG.
AD, (j=1.2...n) and the maximum value therein Vpemk. At T2O, a knock determination is performed. Here, for example, the number CPLS of which VAD≧V r@f is counted. Then, when CPLS is equal to or greater than a predetermined value, it is determined that there is a "knock".

In T2O, the first rounding process of VAD is performed as follows.

VMADj VMADj-1 In step T50, it is determined whether the knock determination period has ended. If YES, the process advances to step T60; if NO, the process returns to step T20. In step T60, the result of the determination as to whether there is a knock or not is sent to the microcomputer 60 through the boat.

In step T70, the second averaging of ■ is performed (details will be described later). In step T80, the knock determination level correction parameters are updated. In step T90, an A/D conversion port for the next knock determination section is selected. The method of correcting the parameter for correcting the knock judgment level is described in detail in Japanese Patent Application Laid-Open No. 126244/1982, and since it is hardly related to the gist of the present invention, the explanation will be omitted here. .

Next, processing contents related to the present invention will be explained in detail. 8th
The figure shows a specific period determination routine. When this routine starts from step SOI, in step 302 the contents of the boat connected to the engine control microcomputer 60 are read. This specific condition is, for example, ■high load area.

■The change in engine speed is greater than a predetermined value.

■Engine speed is within a predetermined range.

Refers to one or a combination of these.

In step S03, it is tested whether the read boat is "°l", and if YES, the process goes to step SO4; if NO, the process goes to step SO7. step s.

In step 4, test whether the specific condition flag is “1゛°” and select YES.
If so, proceed to step SO6; if No, proceed to step SO5. In step SO5, a predetermined value C3 is set in a counter CDLY for determining a predetermined period. Step S
At 06, a specific condition flag is sent. Step SO7
Now clear CDLY. In step SO8, the specific condition flag is reset. CDL in Stembu SO9
Decrease Y by l counts. In step SOA, it is tested whether CDLY is negative, and if YES, the process proceeds to step SOB, and if NO, the process proceeds to step SOC. CDL in step SOB
Guard Y to 0. This routine ends at step SOC. By doing so, CD L Y2O is maintained for a predetermined period after the specific condition is satisfied.

FIG. 9 shows the averaging routine for V. This routine starts from step Sll, and in step S12 CDLY=
O is tested. Here, in the case of YES, the process advances to step S13. Step SV□. −■. In an 13, V mean is updated by □. If No in step S12, step updating is performed. This routine ends in step S15.

FIG. 10 shows the median value update routine. Step 321
This routine starts from step S22. ,〉
■,. will be tested. Here, if YES, the process advances to step S23, and (V,.,=■,.) is converted to DV. Store it in RAM. In step S24, CDL
It is tested whether Y=O, and if YES, step 325
If the answer is No, the process advances to step S26. Step 32
In 5, V2O is DVs.

Increased by □. In step 326, if the answer is No in 2, the process advances to step 327. ■p0□<v,. If YES, the process proceeds to step 328; if NO, the process proceeds to step S2D. In step 328 (V2OV-
--t) to DV, . In step 329, it is tested whether CDLY=O, and if YES, the process advances to step S2A, and if NO, the process advances to step 32B. In step S2C, the contents of DVpsm& are converted to DVs.
Store in o. This routine ends in step S2D.

FIG. 11 shows a knock determination level correction routine. This routine starts from step S31, and in step S32 it is tested whether CDLY=0, and if YES, the process advances to S33.
If No, proceed to S34.

In step S33, the determination level is corrected.

In step S34, the determination level correction parameters are reset. This routine ends at step 335. Here, step 333. S34 is described in detail in Japanese Unexamined Patent Publication No. 126244/1982, and since it has little relevance to the present invention, the explanation will be omitted.

In addition, in the above-mentioned embodiment, ■□. Although the specific engine conditions for increasing the update amounts of and V5(l are the same, specific conditions may be set for each.

For example, ■. The specific conditions for ,1 are: ■High load; ■Engine speed is above a predetermined value. The specific conditions for this may be: (1) high load, (2) engine rotational speed within a predetermined range, and 01727 change in rotational speed less than a predetermined value. Here, ■ and ■ present opposite conditions when V mean is used to create the knock judgment level, and ■. This is because it was assumed that this would be used to correct the knock determination level in a quasi-steady state. That is, ■, □7 needs to follow any transition, but ■,. This is because it is necessary to follow only relatively slow transients.

Moreover, in the embodiment described above, ■□ and ■ only for a predetermined period after the specific condition is satisfied. I increased the update amount, but ■１゜□
Or V. It is also effective to apply the present invention to either one of them.

Furthermore, in the embodiment described above, correction of the knock determination level is prohibited for a predetermined period after the specific condition is satisfied; This is because the response delay is large and there is a strong possibility of incorrect correction. Furthermore, since the response delay of V 111111111 is large, the knock determination is also uncertain. Therefore, it is also effective to prohibit learning of the knock determination level, the amount of retardation, etc. during this period.

Further, in the above embodiment, the predetermined period is defined by the number of ignitions, but it may of course be defined by time.

〔Effect of the invention〕

As described above, in the present invention, the averaging means or the cumulative percentage is
Since the amount of change per one time of at least one of the point detection means is increased, it is possible to improve both the stability and responsiveness of at least one of the average value of the knock intensity value and the cumulative percentage point of the distribution. This has the excellent effect of always being able to achieve accurate knock control.

[Brief explanation of the drawing]

Figure 1 is a claim correspondence diagram of the present invention, Figures 2 (a) to (C
), Figures 3(a), (b) and Figure 4(a), (
b) is various characteristic diagrams for detailed explanation of the present invention, No. 5
The figure is a block diagram showing one embodiment of the device of the present invention, FIG.
8 to 11 are flowcharts for explaining the operation of the apparatus shown in FIG. 5, and FIG. 7 is a diagram showing the knock sensor output waveform and its A/D conversion point. DESCRIPTION OF SYMBOLS 10... Knock sensor, 50... Microcomputer for knock detection, 60... Microcomputer for engine control, 70... Igniter, injector.

Claims (2)

[Claims] (1) A knock sensor for detecting knock of an internal combustion engine, a knock intensity value detection means for detecting a knock intensity value from this knock sensor signal, and an averaging means for averaging the knock intensity value or a knock intensity value detecting means for averaging the knock intensity value. At least one of the cumulative percentage point detecting means for detecting the cumulative percentage point of the distribution, the knock judgment level generating means, the knock judgment means for judging a knock based on the knock judgment level, and the knock judgment result. A knock control device for an internal combustion engine includes a knock control means for controlling knock control factors such as ignition timing and air-fuel ratio according to the engine state, and an engine state detection means for detecting an engine state, and a knock control device for controlling knock control factors such as ignition timing and air-fuel ratio according to the engine state. and update amount setting means for increasing the update amount of at least one of the averaging means or the cumulative percentage point detection means during the predetermined period. A knock control device for an internal combustion engine characterized by: (2) comprising a learning means for learning at least one of the knock determination level or the control amount of the knock control factor; and a learning prohibition means for prohibiting learning of the learning means within a predetermined period by the period determining means. The knock control device for an internal combustion engine according to claim 1.