JPS60243369A - Knocking control method and device for internal-combustion engine - Google Patents

Abstract

PURPOSE:To permit to detect knocking correctly at all times in spite of the manufacturing error or the like of the engine by a method wherein a knock deciding level is corrected so that the shape of distribution of the values of approximate logalithmic transformations of the maximum values in a predetermined division of a knock sensor signal becomes a predetermined shape. CONSTITUTION:The knocking control device decides the knocking by a knock deciding means E based on the output signal of the knock sensor A detecting knocking generated in the internal-combustion engine H. Control values, controlling ignition timing or the knock control factors of a supercharging pressure or the like, are operated by a control value operating means F in accordance with the result of decision and an engine condition signal. In this case, the maximum valve V in a predetermined division of the knock sensor signal is detected by a peak hold means B and the distributing shape of the value LOG(V) of approximate logarithmic transformation of the maximum value V is decided by a distributing shape deciding means C. The knock deciding level of the deciding means E is permitted to be corrected by a correcting means D so that the distributing shape becomes the predetermined shape.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention provides a knock control system that controls knock control factors such as ignition timing, boost pressure, air-fuel ratio, and EOR in accordance with the state of knock occurring in an internal combustion engine. The present invention relates to a control method and device (hereinafter referred to as a knock control system).

[Prior art]

In general, a knock control system is an electrical signal (hereinafter referred to as
When the knock sensor output signal (hereinafter referred to as the knock sensor output signal) exceeds a certain predetermined level (hereinafter referred to as the knock determination level), it is determined that knock has occurred, and the ignition timing is retarded. If no knock is detected, the ignition timing is advanced to keep the ignition timing close to the knock limit, thereby maximizing the fuel efficiency and output characteristics of the engine. (For example, JP-A-56-115861).

In such a knock control system, the knock determination level has extremely important meaning.

If the knock determination level is too high, the ignition timing is advanced because knocking is not detected even though it is occurring, resulting in frequent knocking and even damage to the engine. On the other hand, if the knock detection level is too small,
Even though there is no knock, the ignition timing is retarded and the engine is unable to produce sufficient output.

Conventionally, in order to create an appropriate knock judgment level, for example, the output after passing the knock sensor signal through an integrating circuit,
Constant K carefully adapted to each engine speed
(hereinafter referred to as value) and further adds an offset voltage.

However, because there are manufacturing errors even in the same engine model, the knock detection level may be set to an inappropriate level and accurate knock detection may not be possible even though the I value has been carefully matched. .

As described above, the method of creating a knock determination level in a conventional knock control system requires careful value adaptation and has the problem that accurate knock detection cannot be achieved due to engine manufacturing errors.

[Purpose of the invention]

In view of the above problems, the present invention provides a logarithmically converted value LOG(V) of the maximum value V of the knock sensor output signal during engine combustion.
The knock detection level is automatically corrected based on the experimental fact discovered by the present inventor that the state of occurrence of horn damage can be determined from the distribution shape of the knock detection, thereby eliminating the need for careful 4fi adaptation. The object of the present invention is to provide an unprecedented knocking control method and device that can accurately detect knocking regardless of manufacturing errors in the engine.

[Summary of the invention]

The present invention is broadly divided into the following two parts. The first is a presentation of a basic algorithm for correcting the knock determination level to an appropriate value based on knock signal information. That is, the maximum value of the knock sensor for each cycle is logarithmically converted, and the suitability of the current knock determination level is determined from the distribution shape obtained by sampling a large number of the values.
This provides a basic algorithm that corrects the knock determination level in a desirable direction.

The second objective is to provide a practical method and apparatus for implementing this basic algorithm more easily and at lower cost. That is, based on the conclusions drawn from the basic algorithm, it is not necessary to logarithmically transform the sensor signals, and this algorithm can be replaced by multiplication and division, thus making the processing simpler. be able to. Moreover, this second item includes a method of knowing the multi-cycle distribution shape of the sensor signal without using a large amount of RAM capacity (temporary storage location for data).

First, technical 8! lLL will be explained.

In general, the frequency distribution obtained by sampling a large number of maximum values (peak values) of knock sensor output signals (peak values) for each cylinder during engine combustion in a steady operating state is as shown in FIG. FIG. 3 shows this distribution rewritten as a cumulative distribution with the upper probability as the vertical axis.

By further rewriting such a cumulative distribution as shown in FIG. 4, the inventor discovered the unique properties of the distribution with the maximum value ■.

In FIG. 4, the vertical axis is a value LOG (V) obtained by logarithmically converting the maximum value V of the knock sensor output signal.

The horizontal axis in the figure shows the upper % point of the cumulative distribution (the point where the upper probability is the % value) in correspondence with the ratio U of the deviation to the standard deviation in the normal distribution, using a normal distribution table.

That is, the horizontal axis U is defined by the following equation.

X - μ σ Here, x = LOG (V) μ-ding (i.e., the average value of LOG (V)) σ - σ (X) (i.e., the standard deviation of LOG (V)) This defining formula is shown in Figure 4. Correspondingly, the slope of the straight line in Figure 4 corresponds to σ, and the value of LOG(V) at u=0 is μ
corresponds to To rewrite the cumulative distribution shown in Figure 3 according to the coordinates in Figure 4, for example, if the data at the (upper) 5% point is Vs, find u = 1.6 from the normal distribution table,
The work of plotting the point (1, 6, LOG(Vs)) may be performed for each % point.

In Figure 4, (al and (bl) are L when no knock occurs at all and when knock occurs with a certain frequency, respectively.
It shows the distribution of OG (V). (The fact that al is represented by one straight line means that this distribution follows one normal distribution. Also, (bl is a certain inflection point P
It can be seen that the slope has changed, and this distribution is composed of two different normal distributions.

In addition, the slope of the straight line to the left of the inflection point P at (bl is almost the same as the slope of (al), and the slope of the straight line to the right is about twice as large as the slope of +al. The higher the frequency, the more it moves to the left side of the diagram.From these facts, the distribution to the left of the inflection point P of (al and The distribution to the right of point P is considered to be a distribution unique to when knocking occurs.

The inventor of the present invention has confirmed that such a characteristic is not a phenomenon observed only in a specific engine, but is a phenomenon observed in all engines. As a reference, an example of experimental data is shown in FIG. In addition, the slope of (al) and the slope to the right of the inflection point P that appears when knock occurs do not change much depending on the engine model.The latter slope is about twice as large as the former, so every ignition In the worst case, such as when knocking occurs, the distribution is considered to be a normal distribution with the latter slope, but this distribution cannot be mistaken as the distribution when no knocking occurs.

For the above reasons, the state of knock occurrence can be determined from the distribution of the value LOG (V) obtained by logarithmically transforming the maximum value V of the knock sensor output signal.

Then, by correcting the knock determination level based on the determination result, it is possible to always create an accurate knock determination level.

In this way, it is certainly possible to determine the state of knock occurrence from LOG (V), but it requires expensive equipment such as a logarithmic converter, and it is difficult to understand the distribution of LOG (V). Since it takes a long time and requires a large amount of RAM, further improvements are necessary.

Therefore, the inventor devised the following method as a simple method for correcting the knock determination level by utilizing the characteristics of the distribution shape of LOG (V) described above.

As described above, the distribution when no knock occurs is as shown in FIG. 6 ta+. Here, for example, 1 of the distribution
The values at the 0% point, 50% point, and 90% point are each VIO1V.
50, VSO, LOG (V + o) -L
OG (Vs o) - LOG (Vs o) - LOG
The following relational expression (V9 o) holds true. This formula can be modified as follows.

VIO/VSO=V50/V90 --(11) Also, when knocking occurs, but its frequency is small, it becomes as shown in FIG. 6 (bl), and equation (1) holds true.

However, when knocking occurs frequently, the distribution becomes as shown in Figure 6 tc+, VIO/VSO>VSO/V90 (2
).

In other words, VIO, V'50. VSO relationship is fi1
If it looks like the equation (2), it can be determined that the frequency of knock occurrence is too low; conversely, if it looks like the formula (2), it can be determined that the frequency of knock occurrence is too high.

Next, we will explain how to easily name VIO1VSO1V90. For example, for the maximum value ■ taken this time, V＞
If it is VIO, V10 = V104 with ΔVIO as the set value
-9XΔVIO1V<V,.

Then, V+o=V+o lxΔVIO, so VI
By increasing the ratio of the amount of change in O by nine times, VIO becomes a value that matches the top 10% of the distribution. That is,
If VIO is actually in the top 10%,
The probability that v>v, O is 0, ■, and the probability that V<Vlo is 0.9, so for the captured V, VI
The expected value of the change in O is 9 xO, 1-1xO, 9=O
It becomes stable at the current value. Also, VIO is currently in the top 10
If it is smaller than the value of % points, for example, the value of the top 20% points, the probability that V>V, o is 0.2%, and V
Since the probability that <Vlo is 0.8, the expected value of the amount of change in VIO is 1.8-0.8 = 1, and VIO changes in the direction of increasing and settles at a value in the top 10%.

Similarly, when VIO is larger than the value of the current top 10% point, VI
The expected value of the amount of change in O becomes negative, changes in the direction of decreasing, and settles at the value of the top 10% point.

Also, regarding Vs0, ■〉Vso and V<VSO
The ratio of the amount of change in the case of 1 may be increased to 1, and in the case of VSO, the ratio of the amount of change may be increased to 1/9. Using this method, you can easily find the values at any five points in the distribution.

Moreover, it is better for each amount of change to be a value that corresponds to the knock sensor output rather than to be a constant value.

For example, VSO may be closed as follows. Let ΔVSO be the average of the difference between ■ and Vs0, and at any time, ΔVso-(3XΔ
Vso+1V-Vs0 l)/4 Stop, if v>Vs0, V50=V50+Δ■5o/4
, if V<Vso, then V50=V50−ΔVso/
Set it to 4.

Regarding VIO1V50 and Vso determined in this way, for example, VIO/V5o and Vso are set every 128 cycles.
An accurate knock determination level can be created by calculating the value of s o /V 90, determining the state of knock occurrence based on the magnitude relationship, and correcting the knock determination level.

Also, VIO and Vs0 when the frequency of knock occurrence is small.
The ratio does not change much even if the engine conditions or engine model changes, so the ratio of VIO and Vs0 will be a certain predetermined value (for example, a value that is slightly larger than the ratio of VIO and VSO when no knock occurs). A method of correcting the knock determination level is also effective.

Furthermore, the following method is also effective in order to correct the knock determination level using the characteristics of the distribution shape of LOG (V). This method uses, for example, the median V of the distribution of the maximum value V
Letting SO and a certain constant KO, V is V>KOXVS
This method takes advantage of the fact that the probability that O and the probability that v<vs o / K o are different when a knock occurs and when there is no knock.

The distributions of LOG (V) when no knock occurs at all and when the frequency of knock occurrence is small are shown in Figure 7 (al and (bl), respectively. In these conditions, V
s o / K o < V < K oX V 50
Since the linearity of the distribution of LOG (V) is maintained within the range of , the probability that V > K□XVso and V < V s o
/K.

The probability is equal. In the example shown in the figure, both are 2%.

On the other hand, when knocks occur frequently, the distribution of LOG(V) becomes as shown in FIG. In the example shown in the figure, they are 16% and 2%, respectively. That is, ■〉K
The occurrence state of knocking can be determined based on the probability of O×■5o and the probability of V<V5o/KO.

Also, when no knock occurs, V > K Q
The frequency of VSO does not change much depending on the engine conditions and engine model, so this frequency is set to a certain predetermined value (V>KoXVs when no knock occurs).

It is also effective to correct the knock determination level so that it becomes a value slightly larger than the frequency of .

Furthermore, in the method described above, the knock determination level may be created as, for example, KXV50 by performing four arithmetic operations on the value of a certain percentage point of the ■ distribution.

Now that the purpose, roots, and basic principles of the present invention have been explained above, the configuration of an apparatus for realizing the method of the present invention will be explained next. FIG. 1 is an overall configuration diagram of the system, in which A is a knock sensor that detects non-kinking occurring in the internal combustion engine 11, E is a knock determining means for determining non-king based on the output signal to this knock sensor, and F is the result of this determination and the engine. G is a control value calculation means for calculating a control value for controlling knock control factors such as ignition timing or boost pressure according to a state signal and generating a control signal; B is a driving means for changing the value, and B is the maximum value in a predetermined section of the knock sensor signal.
C is a distribution shape determining means for determining the distribution shape of the approximately logarithmically converted value LOG(V) of this maximum value .E is a knock determining means for determining the distribution shape so that this distribution shape becomes a predetermined shape. This is a determination level correction means for correcting the knock determination level in the knock determination level.

〔Example〕

Hereinafter, the present invention will be explained with reference to embodiments shown in the drawings.

FIG. 8 is a configuration diagram showing an embodiment of the present invention.

In Fig. 8, 1 is a 4-cylinder 4-cycle engine, 2 is an air cleaner, 3 is an air flow meter that detects the intake air amount of the engine and outputs a signal in accordance with this, 4 is a throttle valve, and 5 is a reference crank angle of the engine. This is a distributor that incorporates a reference angle sensor 5A that detects displacement W (for example, top dead center) and a crank angle sensor 5B that generates an output signal for each crank angle of the engine.

6 is a knock sensor for detecting the vibration of the engine block corresponding to the engine knock phenomenon using a piezoelectric element type (piezo element type), electromagnetic type (Magosoto, coil), etc.; 7 is a knock sensor that detects the output of the knock sensor for each cylinder; The peak hold circuit section is shown in bold. 9 is a water temperature sensor that generates a signal corresponding to the engine cooling water temperature, 12 is a fully closed switch (idle switch) that outputs a signal when throttle valve 4 is fully closed, and 13 is a valve that is almost fully open. 14 is a fully open switch (power switch) for outputting a signal when the state is
A/F) is richer or thinner than the stoichiometric air-fuel ratio.
This is a 02 sensor that generates an output signal depending on whether the engine is lean or lean.

8 is an ignition timing control circuit for controlling the ignition timing and air-fuel ratio of the engine according to input/output signal states from each sensor and each switch; 10 is an ignition timing control circuit that receives an ignition timing control signal output from the control circuit 8; An igniter and an ignition coil that cut off power to the ignition coil.

The high voltage generated by the ignition coil passes through the power distribution section of the distributor 5 and ignites the spark plug of a predetermined cylinder at an appropriate time. Reference numeral 11 denotes an injector for injecting fuel into the intake manifold based on the fuel injection time (τ) determined by the control circuit 8.

Next, the detailed configuration of the peak hold circuit section 7 will be explained using FIG. 9. 9, 701 is a filter such as a bunt pass or bypass for selecting and extracting only the knock frequency component from the output signal of the knock sensor 6; 702 is an amplifier; 703
1 is a peak hold circuit that peak-holds the knock sensor signal outputted from 702 based on the cylinder switching signal from the control circuit 8 using, for example, a capacitor.

Next, the detailed configuration and operation of the control circuit 8 will be explained with reference to FIG. In No. 10, 8000 is a central processing unit (CPU) for calculating ignition timing and fuel injection amount.
The 8-bit configuration microbroseso code is used. 8
001 is a read-only storage unit (ROM) for storing control programs and control constants necessary for calculations.
), 8002 is a temporary storage unit (RAM) for temporarily storing calculation data while the CPU 8000 is operating according to a program. 8003 is a waveform shaping circuit for shaping the waveform of the output signal of the crank angle sensor J-5B.

8005 is an interrupt control unit for causing the CPU to perform interrupt processing based on external or internal signals, and 8006 is a CPU
This is a 16-bit timer configured so that the count value increases by one every clock cycle, which is the basic cycle of the U operation. The engine rotation speed and crank angle position are detected by the timer 8006 and the interrupt control unit 8005 in the following manner. That is, each time an interrupt occurs due to the output signal of the reference angle sensor 5A, the CP TJ reads the count value of the timer. Since the count value of the timer increases every clock cycle (for example, 1 μs), by calculating the difference between the count value at the current interrupt and the count value at the previous interrupt, the reference angle sensor signal The time interval, that is, the time required for one revolution of the engine can be measured. In this way, the engine speed can be determined. Further, the crank angle position is determined by the signal from the crank angle sensor 5B, while the crank angle (
For example, the reference angle sensor 5
Based on the top dead center signal of A, the crank angle at that time can be determined in units of 30°A.

This crank angle signal every 30°A is used as a reference point for generating an ignition timing control signal and as a cylinder switching signal for a peak hold circuit.

8007 is a multiplexer that switches multiple analog signals at appropriate times and guides them to an analog-to-digital converter (A/D converter) 8008, and the switching timing is determined by the output port 801.
It is controlled by a control signal output from 1. In this embodiment, the output signal of the knock sensor signal from the peak hold circuit section 7, the intake air amount signal from the air flow meter 3, and the water temperature signal from the water temperature sensor 9 are input as analog signals. 8008 is an A/D converter for converting an analog signal into a digital signal. 8009
is an input port for digital signals, and in this embodiment, the idle signal from the idle switch 12, the power signal from the power switch 13, and the rich lean signal from the 02 sensor 14 are input to this port. 801
0 is an output port for outputting a digital signal.

An ignition timing control signal for the Eve knife 1o, a fuel injection signal for the injector 11, a cylinder switching signal for the peak bold circuit 7, and a control signal for the multiplexer 11 are output from this output port. 8011 is C
It is a PU bus, and control signals and data signals are placed on the CPU bus signal line to control peripheral circuits and send and receive data.

The configuration of the apparatus for realizing the present invention has been described above, and the calculation of ignition timing, knock determination, and correction of the knock determination level will now be described using the flowchart of FIG. 11 as an example.

When the engine is started and an interrupt for ignition timing calculation is performed, the interrupt is started from step 100. In step 101, the engine speed Ne as the engine state
, the basic ignition timing is calculated from information such as the load Q/Ne (Q is the amount of intake air). In step 102, the maximum value V of the knock sensor output signal within a predetermined interval is read for each cylinder. In step 103, the percentage point of the distribution of ■ is calculated. The contents of step 103 will be explained in detail later.

In step 104, it is determined based on the load, etc. whether the engine condition is such that knock control is to be performed, and if no, the process proceeds to step 110, where the ignition timing is adjusted. If YES is determined in step 104, step 10
5, and set the knock determination level Vref for each cylinder, for example, as described above, Vr e f =Kxv5.

From the arithmetic expression:

If V>Vref in step 106, it is determined that a knock has occurred, and the knock flag is set to "H". Depending on this result, a retard amount is calculated in step 107, and ignition timing is calculated in step 108.

Then, the process advances to step 109, and the knock determination level is corrected. The contents of this step 109 will be explained in detail later. Next, the process proceeds to step 110, where the ignition timing is set, and in step 111, the process returns to the main routine.

The overall flow of calculation of ignition timing, knock determination, and correction of knock determination level for carrying out the present invention has been explained above. Next, the flowchart of FIG. The main parts of step 103 of calculating the percentage point of the distribution of ■ and step 109 of correcting the knock determination level will be explained using three embodiments.

The first example shown in FIG. 12 is the 10% point of the distribution of ■,
This is a method of correcting the knock determination level based on the ratio of the values of three percentage points, the 50% point and the 90% point. For convenience,
The portion corresponding to step 103 in FIG. 1I is denoted as 101X, and the portion corresponding to step 109 is denoted as 101-X.

Step 101-X consists of steps X-1 to X-9. In step X-1, the maximum value V read this time and the value of the 10% point of the distribution of
(stored in M), it is determined whether V>V, o. If YES, proceed to step X-2, set VIO-V10+9XV+o as described above, and if No, proceed to step X-3 , VIO-VIO-ΔVIO. By doing this, VIO settles to the value of the top 10% of the distribution. Next, proceed to step X-4, determine V>Vso, and if YES, proceed to step knob X-5,
=VSO+ΔVSO, if NO, step X-6
Proceed to ■5o-V5°-ΔV50. By doing this, V2O settles at the median value of the distribution. Next, the process proceeds to step X-7, and it is determined whether v>v90. If YES, the process proceeds to step X-8, where V2O-V90+ΔV90 is determined, and if No, V2O-V2O9xΔV90 is determined.

By doing this, V2O settles at the top 90% point of the distribution.

In these steps, the value of the % point of the distribution is determined.

Next, step 10 corresponding to step 109 in FIG.
9-X will be explained. Step 109-X consists of steps X-10 to X-15.

In step X-10, it is determined whether the engine is in a steady operating state based on fluctuations in engine speed and load Q/Ne, and if NO, the process proceeds to step X-15. Step X
If YES in -10, the process proceeds to step X-11, and the number of cycles is counted as N=N+1. In step X-12, the steady operating state is set to
It is determined whether or not the sequence continues for O. Here, if YES, proceed to step X-13-2, where V+o/Vso>Vs.

/V Make a 90 judgment. If NO, the process advances to step 110 in FIG. If it is determined as YES in step
If the determination is No, proceed to step X-15, and Vr
Let ef=Vref+A×Δ■. Here, A is set to a value greater than 1.

In this way, the amount of correction A× in the direction of increasing the amount of correction of the knock determination level Vref in the direction of decreasing the amount of correction Δ■
The reason for increasing Δ■ is that when the frequency of knock occurrence is low, the probability of determining YES and the probability of determining NO in step X-13 are equal; This is to ensure that the knock determination level is corrected in the direction of becoming larger (N.
Set to 0.

Next, a second example will be described. This method calculates the median value VSO of the distribution of V as described above, and sets a certain constant K
As O, the probability that V>KOXV50 and V<Vso/K
This is a method of correcting the knock judgment level based on the probability o.

The flowchart in FIG. 13 shows an example of a method for setting the median value of the distribution to 50. This flowchart corresponds to step 103 in FIG. 11, so for convenience, step 103-
It will be expressed as Y. Step 103-Y is step Y
-1 to Y-4.

Step Y-1 is the ■Kichi Vso that is read every cycle.
D=DX3/4+1V-Vsol/4 is determined by the absolute value of the difference between D up to the previous cycle and ■ and VSO in the current cycle. In step Y-2, it is determined whether V>VSO, and if YES, proceed to step Y-3, where V5o=Vso+D/4,
If No, proceed to step Y-4, V50=V50-
Set it as D/4. By doing this, VSO settles at the median value of the ■ distribution.

The reason why we used the average of the absolute values of the difference between ■ and VSO as the amount of change in V in this method is to increase the amount of change in V50 in transient operating conditions so that VSO quickly reaches the median of the distribution. This is to stabilize the VSO with a small amount of change in the regular seat driving state.

Of course, in step 103-X of the first embodiment ■10%
It is effective to use this method when determining VSO and V2O.

Next, correction of the knock determination level by this VSO will be explained using the flowchart of FIG. 14. This part corresponds to step 109 in the flowchart of FIG. 11, and is expressed as step 109-Y for convenience. Step 101-Y consists of steps Y-5 to Y-17.

In step Y-5, it is determined whether the engine is in a steady operating state, and if YES, the cycle number counter is set to N=
Let it be N+1. If No, proceed to step Y-17.
Initialize each counter.

In step Y-7, it is determined that V>KOXVSO, and Y
In the case of ES, the process proceeds to step Y-8, where E-E+1 is set, and the process proceeds to step Y-11. If it is determined No in step Y-7, the process proceeds to step Y-9, where KoXV<
Make a judgment on Vso. Here, in the case of YES, the process proceeds to step Y-10, where F=F+1 is set, and the process proceeds to step Y-11. If NO is determined in step Y-9, the process advances to step Y-11. In step Y-11, it is determined whether the number of cycles N is equal to or greater than a predetermined value N0.
If S, the process goes to step Y-12; if NO, the process goes to step 110 in FIG.

At step Y-12, the rotation E where V>KoXVso
It is determined whether or not E is larger than a predetermined value EMIN. If E is larger than the predetermined value EMIN, the process proceeds to step Y-13; if E is not larger than the predetermined value EMIN, the process proceeds to step Y-17. The determination level is not corrected.In step Y-13, E is set to the predetermined value EM.
It is determined whether or not it is smaller than AX. If YES, the process proceeds to step Y-14; if No, the process proceeds to step Y-15. The value obtained by subtracting F from E in step Y-14 is the predetermined value G.
A determination is made as to whether or not the value is larger than that, and if YES, the process proceeds to step Y-15, and if No, the process proceeds to step Y-16.

In step Y-15, the knock determination level Vref is decreased by a predetermined amount ΔV. In step Y-16-, the knock determination level is increased by a predetermined value Δ■. Next, proceed to step Y-17, N, E.

Change F to O.

A supplementary explanation will be added regarding this second embodiment.

If it is determined NO in step Y-13, step Y
The reason why the knock determination level is made small regardless of the determination of −14 is that when E is very large, knocks occur only frequently. Step Y-14
The reason for setting the threshold value G (G>O) in This is to correct the knock determination level in the direction of increasing it in such a case. Also, a good KO value is about 2.

The third embodiment to be described next is based on the knock determination level Vref.
Create Vr e f = KXVs o, and ■〉Vre
This is a method of correcting the knock determination level by correcting the K value in consideration of the probability f and the probability that V<Vso/. The basic idea is the same as the second embodiment, and is a simplified version. Step 1 in Figure 11
The part corresponding to step 03 is step 101- of the second embodiment.
Since it may be the same as Y, the explanation will be omitted, and the part corresponding to step 109 will be expressed as step 109-Z for convenience.
This will be explained using the flowchart shown in FIG.

Step 109-Z consists of steps Z-6 to z-i3. In step Z-5, it is determined whether the engine is in a steady operating state. If YES, the process proceeds to step Z-6; if NO, the process proceeds to step 110 in FIG. 11. In step Z-6, it is determined whether the knock flag is "H" or not, and it is determined whether Zunawaji, V>KxV5(,). If YES, proceed to step Z-7; If so, the process proceeds to step Z-10. In step Z-7, the value of K is decreased by a predetermined amount Δ1. Next, the process proceeds to step Z-8, where it is determined whether or not K is smaller than the predetermined value KMIN, and Y
If ES, proceed to step Z-9; if No, proceed to step Z-9.
Proceed to step 110 in FIG. K” in step Z-9
“KM I N.

If the determination in step Z-6 is No, the process proceeds to step Z-10, where it is determined that KxV<■5o. If the answer is YES, the process proceeds to step Z-11; if the answer is NO, the process proceeds to step 110 in FIG. K in step Z-11
The value of is increased by a predetermined amount Δ2. Next, proceed to Z-12 and determine whether K is larger than a predetermined value KMAX,
If YES, go to step Z-13; if NO, go to step Z-13.
Proceed to step 110-1 in FIG. In step Z-13, K=KMAX.

A supplementary explanation will be added regarding this third embodiment.

Correction amount Δ1 of K in step Z-7 and step Z-
The relationship between the correction amount Δ2 of K in No. 11 is set so that Δ2 is slightly larger. For example, Δ+ = 1
/32, Δ2 = 1/16. The reason for doing this is that when the frequency of knock occurrence is small, V>KXV50.
The probability of V<Vso/ is equal to the probability that V<Vso/, and in such a case, the knock determination level is corrected in the direction of increasing it.

Note that the initial value of K is preferably about 2.

〔Effect of the invention〕

As described above in detail, the present invention corrects the knock determination level by utilizing the very nature of vibrations associated with engine combustion, so it does not require careful value adaptation, and
It is possible to provide a new non-king control method and device that can always accurately detect knocking regardless of manufacturing errors in the engine.

Furthermore, when carrying out the present invention, it is only necessary to obtain the maximum value of the knock sensor output signal during engine combustion;
There is no need to significantly change the conventional knock control system, and it is only necessary to slightly change the processing content of the ignition timing control means.

[Brief explanation of drawings]

Fig. 1 is an overall configuration diagram to clearly show the structure of the device according to the present invention, Fig. 2 is a frequency distribution diagram of the maximum value ■ of the knock sensor output signal, Fig. 3 is a cumulative distribution diagram of ■, and Fig. 4 is a diagram showing the relationship between the ratio U of deviation and standard deviation in a normal distribution and LOG (V), Figure 5 is a characteristic diagram showing an example of experimental data of LOG (V), and Figures 6 and 7 are diagrams showing the relationship between the ratio U of deviation and standard deviation in a normal distribution and LOG (V). FIG. 8 is a diagram showing an embodiment of the apparatus for carrying out the present invention, FIG. 9 is a block diagram of the peak hold circuit section, and FIG. 10 is a diagram showing the control circuit in FIG. 8. 11 is a flowchart showing the ignition timing calculation, knock determination, and knock determination level correction procedure in the present invention, and FIG. 12 shows a first embodiment of steps 103 and 109 in FIG. 11. 13 and 14 are flowcharts showing a second embodiment of steps 103 and 109 in FIG. 11, and FIG. 15 is a flowchart showing a third embodiment of step 109 in FIG. be. 1...Engine, 5...Distributor, 6.
...Knock sensor, 7...Peak hold circuit section, 8
... Ignition timing control circuit, 10... Igniter and ignition coil, 703... Peak hold circuit. 8000...Central processing unit, 8001...RO
M. 8002...RAM. Representative Patent Attorney Takashi Okabe Figure 1 Figure 2 Figure 3 Figure 4 Distribution of 1 flight 1a (%) Figure 5 9 [195908070GO5040:M) 20 To
5 2 Detective vjp! (ㅅ・)Figure 6(a) (bl - Vessel---Shizuku---Ikei---/ Figure 13 Figure 15 Figure 14

Claims (8)

[Claims] (1) To detect knocking occurring in the internal combustion engine using a knock sensor, determine knocking based on the output signal of this knock sensor, and control knock control factors such as ignition timing or boost pressure according to the determination result. In the knock control method for an internal combustion engine, which generates a control signal and controls the knock control factor according to the control signal, the distribution of approximately logarithmically converted values LOG (V) of the maximum value V in a predetermined section of the knock sensor signal. A non-king control method for an internal combustion engine, characterized by correcting a knock determination level so that the shape becomes a predetermined shape. (2) The internal combustion engine according to claim 1, characterized in that the knock determination level is corrected based on the magnitude relationship of the ratio of the values of at least three or more percentage points of the distribution of the maximum value (■). Knocking control method. (3) The internal combustion according to claim 1, wherein the knock determination level is corrected so that a ratio of values of at least two or more percentage points of the distribution of the maximum value (■) becomes a predetermined value. Engine knocking control method. (4) A patent characterized in that the knock determination level is corrected so that the frequency at which the maximum value ■ exceeds a value obtained by multiplying the value of at least one percentage point of the distribution by a certain constant becomes a predetermined value. A knocking control method for an internal combustion engine according to claim 1. (5) The knock determination level is corrected based on the magnitude relationship between the maximum value ■ and at least two or more values obtained by performing four arithmetic operations on a predetermined value of the median value of the distribution. 2. The knocking control method for an internal combustion engine according to item 1. (6) The knocking control method for an internal combustion engine according to any one of claims 1 to 5, characterized in that the knock determination level is corrected when the internal combustion engine is in a steady operating state. (7) The knock determination level is created by performing four arithmetic operations on a predetermined value on the value of at least one percentage point of the distribution of the maximum value V.
The non-king control method for an internal combustion engine according to any one of paragraphs. (8) A knock sensor that detects knocking that occurs in an internal combustion engine, and a system that determines knocking based on the output signal of this knock sensor and controls knock control factors such as ignition timing or boost pressure according to the determination result. In a non-king control device for an internal combustion engine, comprising a control signal generating means for generating a control signal, and a driving means for changing the value of the knock control factor according to the control signal, the maximum value V of the knock sensor signal in a predetermined section a peak hold means for detecting the maximum value V, and an approximately logarithmically converted value LOG (
V) A knocking control device for an internal combustion engine, comprising: a distribution shape determining means for determining the distribution shape of the noise; and a determination level correcting means for correcting a knock determination level so that the distribution shape becomes a predetermined shape.