JP3093467B2 - Engine knock detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a knock detecting device for an engine, and more particularly, to a device for detecting a knock intensity by analyzing a frequency of a knock sensor signal.

[0002]

2. Description of the Related Art When knock occurs in an engine, vibration having a plurality of resonance frequencies specific to each engine is generated. It is known that the magnitude of knock is general if it is expressed in terms of total vibration energy, and the total vibration energy due to knock can be obtained by calculating the sum of the resonance frequency components. As in the publication, the sum of a predetermined frequency of the frequency analysis result output is obtained.

[0003] By the way, as shown in FIG.
Frequency analysis of the signal from _Ten~ Ρ₃₀Like
The frequency spectrum intensity of the number increases. And ρ_Ten~
ρ₃₀Of the spectral intensity from
Because it fluctuates depending on the generated cylinder, etc.
If only the wavenumber band is detected and knock determination is performed, knock
In some cases, it cannot be detected (FIGS. 18 and 19). There
The frequency analysis of the knock sensor output (FFT, WFT)
Etc.) and multiple filters (analog, digital)
And add the spectral intensities in multiple frequency bands
Knock determination was made based on the result.

Here, the energy generated by knocking can be regarded as the sum of the spectral intensities of the resonance frequencies of the respective vibration modes, so that the above-described determination of the knocking intensity by the addition is effective.

That is, it is natural that better control (rapid and safe) can be performed if the knock intensity can be determined, rather than simple knock determination.

[0006]

However, ρ _10-
[rho ₃₀ of spectral intensity in Mode S / N (knock there during / noise), because it depends on the mode, the addition of a simple spectral strength, despite a certain one mode is determined that there is knock, another Since the output in the mode is low, there is no knock when comprehensively judged, and no detection is possible.

The converse is also conceivable. In order to solve the above problems, an object of the present invention is to perform more accurate knock detection.

[0008]

SUMMARY OF THE INVENTION Therefore, the present invention provides a knock sensor for detecting knock of an engine, an A / D converter for A / D converting a signal from the knock sensor at predetermined time intervals, and an A / D converter. Frequency analysis means for frequency-analyzing the D-converted data, knock determination means for performing knock determination of the engine based on a plurality of frequency components obtained from the frequency analysis result, and knock determination among the plurality of frequency components And a frequency component selecting means for selecting a frequency component according to an operation state of the engine.

[0009]

Thus, the knock sensor detects the knock of the engine, and outputs a signal from the knock sensor to the AD.
A / D conversion is performed at predetermined time intervals by the conversion means. And
The A / D-converted data is subjected to frequency analysis by frequency analysis means, and knock determination of the engine is performed by knock determination means based on a plurality of frequency components obtained from the frequency analysis result. Further, a frequency component used for knock determination among a plurality of frequency components obtained from the frequency analysis result is selected by the frequency component selection means according to the operating state of the engine.

[0010]

According to the present invention, there is an excellent effect that even if the knock frequency component fluctuates according to the engine operating state, the knock can be accurately determined under all operating conditions without malfunction.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, the knock phenomenon will be described. Knock refers to a phenomenon in which unburned gas in a cylinder is compressed by combustion gas, self-ignites, and rapidly burns, causing resonance in the cylinder. If the knock is controlled at a minute level, fuel economy can be improved without damaging the engine. As disclosed in the embodiment of JP-A-3-47449, the above-mentioned resonance has a natural frequency determined by the bore diameter and the speed of sound of the engine, the order in the cylinder radial direction is n, and the order in the circumferential direction is m. When the resonance vibration mode is P
If it is set to nm, for example, it is predicted that a knock component appears at a frequency as shown in FIG.

However, when the knocking frequency is analyzed in an actual engine, a knocking component is not always generated at the frequency shown in FIG. 3B. FIG. 3A shows frequency analysis results when a knock is generated for one ignition of a signal from a knock sensor attached to an engine block (not shown) and when there is no knock. The vibration modes corresponding to ρ ₁₀ and ρ ₂₀ match the values expected from FIG. 3B, but ρ ₀₁ and ρ ₃₀ are 14.6 KHz, 16.
15.5KHz, 1 though predicted to be 0KHz
A knock component occurs at 6.5 KHz. ρ ₁₀ , ρ ₂₀
In other ignition cycles of the mode, it is necessary to search for the knock generation frequency for each ignition because the ignition timing varies similarly.

Next, a knock determination method will be described. The following findings were obtained from engine combustion experiments. That is, in the vibration sensor attached to the block, the output of each knock occurrence frequency is different every time knock occurs. For this reason, it is desirable to determine a knock by setting a knock detection threshold for each knock occurrence frequency.

Following the knock determination method, a specific configuration will be described. 1 and 2 are a functional block diagram and an overall configuration diagram showing an embodiment of the present invention. In these figures, reference numerals 1 and 8 denote knock signal detecting means and a knock sensor, which convert vibration of the engine body into an electric signal. In the present invention, it is necessary to use a non-resonant sensor for performing frequency analysis. Reference numeral 2 denotes a pre-processing means, a receiving circuit 9 for adjusting the impedance of the sensor signal, a low-pass filter (hereinafter abbreviated as LPF) 10 for preventing aliasing noise, and a high-pass filter for removing a low frequency which is not related to knock. (Hereinafter abbreviated as HPF) 11. LPF
10 is configured to pass a signal having a frequency of 20 KHz or less, and HPF 11 is configured to pass a signal having a frequency of 1 KHz or more.

Reference numerals 3 and 12 denote gain switching means and a gain switching device for adjusting an input signal to an appropriate magnitude. Here, an A / D converter with high resolution may be used without using the gain switching means 3 and 12, but it is expensive. Subsequent FF
In order to use T, this embodiment provides a dynamic range wider than that of the conventional one, equivalent to a 16-bit A / D converter. Reference numeral 13 denotes a digital signal processor (hereinafter abbreviated as DSP), which can process multiplication at a higher speed than a microcomputer. A / D converters 4 and 14 are included in the DSP 13.
And a parallel I / O (hereinafter abbreviated as PiO) 15. P
The iO 15 selects a gain and communicates with the host CPU.

Reference numerals 5, 6A, and 6 denote frequency analysis means, knock frequency search means, and knock determination means constituted by software in the DSP 13, which respectively perform FFT (frequency analysis by fast Fourier transform) and knock generation from the FFT result. A search for each frequency ignition and a knock determination are performed. 16
Is a host CPU, which controls the ignition timing to be output to the ignition device 17 and the fuel injection amount (not shown) based on the result of the presence or absence of knock from the DSP 13.

Next, the operation of the DSP 13 will be described with reference to the timing chart of FIG. A in FIG. 4 is a ATDC-10 ° CA falling each cylinder of the signal in indicates the reference position signal, whereby the reference position interrupt FIG. 6 which will be described later in the period a to time t ₁ in FIG. 4 The processing is executed. B in FIG. 4 shows a knock determination section, and b and c in FIG. 4 show a period of a timer 1 interrupt process of FIG. 7 described later, which is executed from time t ₂ and t ₃ . D in FIG.
Indicates the period of the timer 2 interrupt processing of FIG. 8 to be described later is executed from time t _2. FIG. 4e shows a processing period such as FFT executed in the main routine.

First, at the falling edge of the reference position signal at time t _{1 in} FIG. 4, reference position interruption processing a is executed, and the start time of the knock determination section is set in the timer 1 as indicated by f. As a result, the timer 1 interrupt process b is performed becomes a determination section start time of the time t _2. This allows
The end time of the knock determination section is reset in the timer 2 as indicated by g, and the timer 2 interrupt processing d is executed to start AD conversion of the knock signal. This process is performed once every 20 μsec using the timer 2 interrupt.

In principle, the FFT can handle only 2 ⁿ data of a predetermined number. Therefore, here, FFT is performed in units of 2 ⁷ = 128. The A / D converted values are accumulated in a time series, and when the number reaches 128, the FF in FIG.
In the section T1, the knock signal data AD-converted in the section AD1 is subjected to FFT. At the same time, the AD conversion is performed in the section AD2.
Following AD3. Each section AD1, AD2, AD3 is divided into 128 sections of AD conversion data in each section AD1, AD2.
The A / D conversion operation is performed identically and continuously during the knock determination section, and each A / D conversion data is accumulated in a time series. Section A
Similarly, when the number of data reaches 128 in D2, the section FFT is performed.
2 performs the second FFT. This second FFT result is added to the first result.

When the knock determination section ends, after the FFT is completed in the main routine, a knock occurrence frequency is searched for each ignition based on the addition result of each FFT in one knock determination section to determine knock intensity. Then, the result is sent to the host CPU 16. Note that the knock determination section usually does not end at the same time as the end of the 128-point A / D conversion.
Abort D conversion. In this case, the remaining A / D conversion points are 128 bits on the assumption that the A / D conversion input is 0.
Perform a point FFT.

When the knock detection section ends before the previous FFT is completed, the next FFT is executed after waiting for the end of the previous FFT. Multiple times of FF as above
With the configuration in which the result of T is added, the knock determination section can be set freely. At the same time, when the rotation speed is low, the frequency can be analyzed with a smaller amount of calculation than the frequency analysis of the entire section by one FFT.

FIGS. 5 to 15 are flowcharts showing the flow of the program of the DSP 13. First, the program is started from step M10 in FIG. 5 simultaneously with turning on the power. Steps M20 and M30 are various initial settings, and are executed only once at first.

Next, the reference position interrupt program shown in M40 of FIG. 6 by angle interruption at the reference position at the time t ₁ in FIG. 4 is started. Here, the reference position indicates the crank angle of the engine, which is 10 ° crank angle (CA) before the top dead center (BTDC) of each cylinder, and is sent from the host CPU 16. First, in step M50, the cylinder is determined, and a value corresponding to the distance from the currently ignited cylinder to the knock sensor is output. This signal is also a signal sent from the host CPU 16 and is 1 only for the first (# 1) cylinder, and the other cylinders are calculated using a counter. Step M
Reference numeral 60 denotes a rotation speed calculation portion, which calculates the rotation speed based on the time required from the previous reference signal to the current reference signal. Step M7
0 performs input gain setting, fail determination, timer setting, and the like as preprocessing. In step M80, a timer 1 is set to wait for a crank angle set in advance according to the engine speed. Here, the knock usually occurs approximately 15 ° CA to AT after the top dead center (ATDC).
Since DC is about 70 ° CA, in step M80, a time corresponding to a value of ADTC about 10 ° CA slightly earlier is set.

When the time set at step M80 (t _{2 in} FIG. 4) comes, the timer 1 indicated by M90 in FIG.
The interrupt program is started. First, step M91
In step M92, it is determined that a knock determination section is started because a knock determination end flag described later is 1.
Then, the knock determination section end flag is set to 0. Then M
Proceeding to 93, the timer 1 is reset to a time corresponding to a value of about 70 ° CA of ATDC, and then proceeding to step M94, the timer 2 for starting A / D conversion is set and activated. Thereafter, the process proceeds to step M95 to start the main routine.

When the reset time of the timer 1 is reached (t _{3 in} FIG. 4), the timer 1 interrupt program indicated by M90 in FIG. 7 is started again. This time, in step M91, since the knock determination end flag is 0, it is determined that the section is the knock determination end section, and the routine proceeds to step M96. After the knock determination section end flag is set to 1, the routine proceeds to step M97 and the timer 2 is reset. The operation is stopped to end the AD conversion.

Next, the timer 2 interrupt M97 shown in FIG. 8 will be described. This timer 2 interrupt is 20 μsec
It is executed every time. First, in step M99, the A / D conversion of the knock signal by the A / D converter 14 is started. Then, the process proceeds to step M100, in which the value A / D converted by the A / D converter 14 is taken into the DSP 13. , Step M10
The program proceeds to 1 to store and store the fetched A / D converted values in a RAM (not shown) in the DSP 13 in time series.

FIG. 9 shows the pre-processing of step M70 in FIG. 6. First, in step M71, the gain of the gain switch 12 is adjusted so that the input signal has an appropriate magnitude. Next, the routine proceeds to step M72, in which a failure of knock sensor 8 is determined. As shown in FIG. 10, in this fail determination method, in step M721, a predetermined value is obtained by comparing a value obtained by smoothing the value of PALL detailed in FIG. If the positive value is smaller than the predetermined value, the flow advances to step M722 to supply a sensor fail output to the host CPU 16.

FIG. 11 shows a main routine started in step M94 of FIG.
02, it is determined whether the knock determination section end flag is 1; if the knock determination section end flag is not 1, step M10
3 to determine whether the 128-point A / D conversion has been completed. If not completed, return to step M102. If completed, proceed to step 104 to execute FFT and add the result to the previous time. Add to the result.

Next, when the knock determination section end flag is 1 in step M102, the process proceeds to step M105, in which the value of the portion less than 128 during the A / D conversion at that time is set to a predetermined value = 0, and then the process proceeds to step M106. The FFT is executed to proceed to and the result is added to the previous addition result.

Then, in the next step M107, post-processing such as searching for a knock occurrence frequency for each ignition is performed based on the addition result of each FFT in one knock determination section, and then in step M108, one ignition is performed. The knock strength is determined for each time.

FIG. 12 shows the step M107 of FIG. 11 in more detail.
From the frequency-spectral intensity characteristic corresponding to FIG. 3A obtained based on the result of addition of the FT, step M110
1 to M1105, each knocking frequency ρ ₁₀ , ρ ₂₀ , ρ
_01, [rho _30, by executing each of the peak value search and each output Get in [rho ₁₁ mode, output corresponding to each of these knocking frequencies for each ignition is searched each knock frequency is calculated. In the next step M1106, the total power of the knock signal is calculated.

Referring to FIG. 13, step M1101 in FIG.
Will be described in more detail. Note that step M11 in FIG.
02 to M1105 are basically the same as FIG. FIG.
FIG. 16 is a simplified illustration of the addition result of each FFT corresponding to the main part of the characteristic of FIG. 16A, and a method of peak value search (knock occurrence frequency search) and output calculation will be described on behalf of this characteristic. .

[0033] First of all, read from the ROM (not shown) the center frequency f ₁₀ in the ρ ₁₀ mode at step M131. Here, as shown in FIG. 20, each of the vibration modes [rho ₁₀ ~
Each center frequency f ₁₀ for each [rho ₁₁ is stored in the ROM. Upper and lower limit frequency f _u in [rho ₁₀ mode in the next step M132, calculates a f _l. Here, the upper and lower limit frequency f
_u, f _l is obtained by adding and subtracting the resonant mode frequency set in advance in anticipation of predetermined margin to the center frequency f ₁₀ of the (knocking frequency) [rho ₁₀ (e.g., 2 KHz).

In the next step M133, the upper and lower limit frequencies f _L,
After finding the maximum value f _MAX of the spectral intensity within _fu ,
Obtaining the sum P ₁₀ of intensity at f _MAX ± 1172Hz at step M134. Next, step M13
5, a value MD ₁₀ obtained by normalizing the sum P ₁₀ of the spectrum intensities.
Ask for. That is, the sum of the spectral intensity obtained by multiplying the knocking determination value K ₁₀ at a distance and total data number and [rho ₁₀ mode from the corresponding cylinder to knock sensor 8 values P
₁₀ by dividing the determined value MD ₁₀ that has been normalized. Here, as shown in FIG. 21, the knock determination value for each vibration mode ρ ₁₀ ~ρ ₁₁ it is previously stored in a not shown ROM.

Referring to FIG. 14, step M1106 in FIG.
Will be described in more detail. In step M141, the sum PALL of all the spectral intensities of 5 to 20 KHz is obtained from the frequency-spectral intensity characteristic corresponding to FIG. 3A obtained based on the addition result of each FFT in one knock determination section.

FIG. 15 shows the knock determination step M1 of FIG.
08 in more detail. First, step M151
After prohibiting the interruption in step M152, it is determined whether or not the engine speed is equal to or more than 6000 rpm. Go ahead, M
The D ₁₀ and MD ₂₀ is cleared to 0, and N = 3. Also if it is determined not to be the first cylinder or fourth cylinder in step M153, it clears the MD ₂₀ to 0 at step M156, an N = 4. Further, when the result in step M152 is negative, N = 5 in step M158.

The knock frequency varies depending on the position of the knock sensor 8 and the engine speed. For example, in an in-line four-cylinder engine, one knock sensor 8 is provided in a cylinder block between the second and third cylinders. If only fixed,
At high engine speeds, the spectrum intensity of the high frequency band becomes stronger in the cylinders (first and fourth cylinders) farther from the knock sensor 8, and thus the frequency band selected according to the engine operating state is changed. is there.

[0038] Next, in step M159, MD ₁₀ ~M
Seeking MD _ALL a value obtained by adding all the D ₁₁ is divided by N,
Store in RAM. In the next step M160, knock intensity KNK is calculated from the TKNK table shown in FIG. 17 using MD _ALL obtained in step M159.

Then, steps M102 to M1 in FIG.
Reference numeral 06 corresponds to the frequency analysis means of the present invention.
Steps M152 to M158 correspond to frequency component selection means of the present invention, and step M108 in FIG. 11 corresponds to knock determination means of the present invention.

In the embodiment described above, FIG.
In the steps M154 and M156, the MD ₁₀ value and the MD ₂₀ value are cleared to 0, but a predetermined value close to 0 may be substituted instead of clearing to 0.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing one embodiment of the present invention.

FIG. 2 is an overall configuration diagram showing one embodiment of the present invention.

3A is a diagram showing a frequency-spectrum intensity characteristic of a knock sensor signal, and FIG. 3B is a diagram showing a knock occurrence frequency analysis of an engine.

FIG. 4 is a timing chart of the embodiment.

FIG. 5 is a flowchart of an initial setting in the embodiment.

FIG. 6 is a flowchart of a reference position interrupt in the embodiment.

FIG. 7 is a flowchart of a timer 1 interrupt in the embodiment.

FIG. 8 is a flowchart of a timer 2 interrupt in the embodiment.

FIG. 9 is a flowchart of preprocessing in the embodiment.

FIG. 10 is a flowchart of a sensor failure determination in the embodiment.

FIG. 11 is a flowchart of a main routine in the embodiment.

FIG. 12 is a flowchart of post-processing in the embodiment.

13 is a flowchart of the search ₁₀ mode peak value ρ in the above embodiment.

FIG. 14 is a flowchart of a total power calculation in the embodiment.

FIG. 15 is a flowchart of knock determination in the embodiment.

FIG. 16 is a characteristic diagram for explaining an operation of a peak value search in the embodiment.

FIG. 17 is a knock strength calculation table (TKNK) in the embodiment.

FIGS. 18 (a) and (b) are frequency-spectrum intensity characteristic diagrams in different operating states.

FIGS. 19 (a), (b), and (c) are frequency-spectrum intensity characteristic diagrams at different rotational speeds.

FIG. 20 is a diagram showing a center frequency value stored in R0M.

FIG. 21 is a diagram showing a knock determination value stored in R0M.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Knock signal detection means 4 A / D conversion means 5 Frequency analysis means 6 Knock determination means 6A Knock frequency search means 7 Ignition timing control means 8 Knock sensor 13 Digital signal processor 14 A / D converter 16 Host CPU 17 Ignition device

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Yoshiki Chujo 1 Toyota Town, Toyota City, Aichi Prefecture Inside Toyota Motor Corporation (72) Inventor Masato Sabashi 1 Toyota Town, Toyota City, Aichi Prefecture Inside Toyota Motor Corporation ( 56) References JP-A-5-332192 (JP, A) JP-A-4-339157 (JP, A) JP-A-3-47449 (JP, A) JP-A-4-76249 (JP, A) JP Hei 3-267547 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) F02D 45/00 368 F02P 5/152

Claims (1)

(57) [Claims] 1. A knock sensor for detecting knock of an engine, A / D conversion means for A / D converting a signal from the knock sensor at predetermined time intervals, and frequency analysis of the A / D converted data. Frequency analysis means, knock determination means for performing knock determination of the engine based on a plurality of frequency components obtained from the frequency analysis result, and a frequency component used for knock determination among the plurality of frequency components is selected according to an operating state of the engine. A knock detection device for an engine.