JP2570912B2 - Misfire detection method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting a misfire in a multi-cylinder engine, and more particularly, to calculating a time width in a set crank angle range for each cylinder based on crank angle information, and responding to the fluctuation of the time width. And a misfire detection method for obtaining misfire information.

[0002]

2. Description of the Related Art An engine requires a fuel supply amount and an intake air amount corresponding to an operation state of a combustion chamber thereof.
It is necessary to perform ignition or ignition (hereinafter referred to as ignition or the like) at the ignition timing or ignition timing (hereinafter referred to as ignition timing or the like) according to the operation state. When the fuel supply amount, the intake amount, the ignition timing, and the like maintain a steady state, the output amount of each cylinder of the engine continues uniformly, and the change in the adjacent engine rotation time width is almost unchanged. Does not occur. On the other hand, when an inappropriate fuel supply amount, intake air amount and ignition timing are set for the engine operation state, or when the engine body itself becomes more than one, at least some of the cylinders out of all the cylinders are used. Misfires, stops producing power and, in the worst case, shuts off the engine.

[0003] Therefore, it is conventionally known to detect a misfire of an engine and display a warning. For example, in a six-cylinder engine, misfire determination is performed by taking in crank angle information as shown in FIG. Here, the cylinders along a predetermined ignition sequence are sequentially ignited at every crank angle of 120 ° to generate an output. In this case, in the cycle T _n-1 of the first cylinder # 1, first, the elapsed time T _cn of the first crank angle region c1 from 75 ° before the top dead center to 5 ° before the top dead center of the first cylinder # 1. _-1 and the elapsed time ratio T _en-1 / T _cn-1 between the first crank angle region c1 and the second crank angle region e1 before the first crank angle region c2 of the second cylinder # 2. It is calculated as the previous time ratio. Further, the first crank angle region c3 of the third cylinder # 3, which follows the first crank angle region c2 of the second cylinder # 2 and the first crank angle region c2 and reaches the top dead center next to the second cylinder # 2. calculating the ratio of the elapsed time between the second crank angle range e2 and before the rear elapsed time ratio T _en / T _cn.

Then, the following equation (1), that is, the previous elapsed time ratio is subtracted from the later elapsed time ratio, and the value is added to the first cylinder # 1.
Of the reciprocal of the cycle T _n-1 of
Dn was corrected to the dimension of the rotational acceleration), the first variation value ΔDn
(♯1) is calculated. ΔDn (♯1) = (1 / T n-1 2) × (T en / T cn -T en-1 / T cn-1) (1) After this change value [Delta] D _n (# 1) is pre-experimental The misfire is determined based on whether or not the value exceeds the threshold value set in (1).

[0005] Similarly, also in the following second cylinder ♯2 period T _n of sequentially before the elapsed time ratio T _en / T _cn, calculates the post-elapsed time ratio _{T en + 1 / T cn +} 1, on the The second variation value ΔD (♯
2) is calculated, and misfire is determined based on the second variation value ΔD _n (♯2).

FIG. 12 shows a variation value ΔD _{n that} changes with time.
A chart is shown using the above data as an example.

[0007]

However, once a misfire has occurred (shown by a mark に in the figure), the level of such a fluctuation value ΔD _n rapidly increases. However, after this misfire, the value often increases rapidly even though the misfire has not occurred (indicated by the symbol ▽ in the figure).
Proliferation of variation value [Delta] D _n after such misfire is primarily is estimated that based on the torsional stiffness of the crankshaft.

[0008] Thus, the rapid increase of the variation value [Delta] D _n after misfire on the misfire detection, there is a problem of lowering the detection accuracy. An object of the present invention is to provide a misfire detection method capable of improving misfire detection accuracy.

[0009]

In order to achieve the above-mentioned object, a method according to the present invention comprises the steps of: setting a first cylinder before a top dead center;
Setting the second cylinder following the first crank angle range and reaching the top dead center next to the above-described one cylinder.
The respective elapsed time ratios with respect to the set second crank angle region before the crank angle region are calculated as the previous elapsed time ratios, and the following first crank angle region and the same set first crank angle region of the second cylinder are followed. The respective elapsed time ratios of the third cylinder reaching the top dead center next to the second cylinder and the set second crank angle region before the set first crank angle region are calculated as post-elapsed time ratios. Subtracts the previous elapsed time ratio from the post-elapsed time ratio to calculate a first variation value, then calculates a second variation value according to the first variation value, and then calculates a second variation value from the first variation value. The method is characterized in that a fluctuation width is calculated by subtracting the fluctuation value, the fluctuation width is multiplied by the first fluctuation value to obtain a misfire degree indication value, and a misfire determination is made based on the misfire degree indication value.

[0010]

The ratio of each elapsed time between the set first crank angle region and the set second crank angle region of the first cylinder is calculated as the pre-elapsed time ratio, and the ratio of the second cylinder is set as the post-elapsed time ratio. First
The ratio of each elapsed time between the crank angle region and the set second crank angle region is calculated, the first variation value is obtained by subtracting the previous elapsed time ratio from the post elapsed time ratio, and the second variation value is obtained as the first variation value Is calculated according to the following formula, and the variation range is calculated by subtracting the second variation value from the first variation value. The variation range is multiplied by the first variation value to obtain a misfire degree indication value. Since misfire judgment is based on
The misfire degree instruction value more accurately includes misfire information as compared with each variation value.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A misfire detection method according to the present invention will be described below with reference to the block diagram of FIG. In this misfire detection method, the calculation process of the first variation value ΔD _n and the second variation value ΔD
_{n + 1} of the calculation process and calculation of the misfire index indicated value I _MF and the misfire determination process based on the misfire index instruction value I _MF is performed in this order. Note that the calculation processing of the first and second fluctuation values ΔD _n and ΔD _{n + 1} is performed in the same manner as in the related art, and therefore, redundant description will be omitted with reference to FIG.

The calculation process of the first fluctuation value ΔD _n is performed in the first cylinder ♯
It is performed in _one cycle T _n-1 . Here, the first cylinder # 1
Is calculated as the elapsed time ratio T _en-1 / T _cn-1 between the elapsed time T _cn-1 of the first crank angle region c1 and the second crank angle region e1. Further calculates a first crank angle range c2 of the second cylinder ♯2 the ratio of the elapsed time between the second crank angle range e2 as a post elapsed time ratio T _en / T _cn.

Then, the first fluctuation value ΔD _n is calculated based on the above equation (1). _{ΔD n = (1 / T n} -1 2) × (T en / T cn -T en-1 / T cn-1) ··· (1) Similarly, the second calculation process of the variation value [Delta] D _{n + 1} is 2nd cylinder $ 2
It performed in the period T _n. Here, the elapsed time T _cn of the second crank angle range c2 to top dead center 5 ° from top dead center 75 ° for the second cylinder # 2, 3 cylinder with subsequent in the second crank angle range c2 The ratio of the elapsed time to the second crank angle region e2 before the first crank angle region c3 of # 3 is the previous elapsed time ratio T
_It is calculated as _en / _Tcn . Furthermore, the first crank angle range c3 of the third cylinder # 3 and the first crank angle range of the fourth cylinder # 4 following the first crank angle range c3 and reaching the top dead center next to the third cylinder # 3.
The ratio of the elapsed time before the crank angle region c4 to the second crank angle region e3 is calculated as a post-elapsed time ratio T _{en + 1} / T _{cn + 1} .

Then, according to the equation (1), the following (1) ′
Based on the elapsed time ratio T _{en + 1} / T _{cn + 1} and the previous elapsed time ratio T
_A second variation value ΔD _{n + 1} corresponding to _en / T _cn is calculated. ΔD _{n + 1} = (1 / T _n ² ) × (T _{en + 1} / T _{cn + 1} −T _en / T _cn ) (1) ′ First variation value ΔD _n and second variation value ΔD _{n +1} is calculated, and then the process _{proceeds to} the calculation of the misfire degree instruction value IMF.

Here, the first variation value ΔD _n and the second variation value ΔD _{n + 1} are substituted into the following equation (2). Then, from the first change value [Delta] D _n by subtracting a second variation value [Delta] D n _{+ 1} to calculate the variation range, and calculates the misfire degree instruction value I _MF in the variation range by multiplying the first change value [Delta] D _n . I _MF = ΔD _n × (ΔD _n −ΔD _{n + 1} ) (2) The first fluctuation value ΔD _n is multiplied by the fluctuation width (ΔD _n −ΔD _{n + 1} ), whereby the front and rear fluctuations are obtained. It acts so that the value difference is further enlarged.

[0016] After that, the misfire determination process based on the misfire index instruction value I _MF is misfire degree instruction value I _MF sequentially calculated is determined whether a misfire or not according to a predetermined determination condition. If misfire is considered, misfire information is output at that time. As a determination condition in this case, a predetermined threshold value for discriminating the average fluctuation value ΔD _n level at the time of misfire and the normal fluctuation value ΔD _n level is used, and whether or not the threshold value is exceeded is determined. The misfire may be determined based on the result. According to such a misfire detection method, as compared with a conventional case where each variation value ΔD _n calculated sequentially is compared with a misfire determination threshold value to determine whether or not a misfire has occurred, the threshold value is set to a normal variation value. It can be taken sufficiently larger than the level of ΔD _n and the detection accuracy is improved.

Next, a misfire detection apparatus and data obtained by the present inventor using the method of the present invention in an engine control system of an engine will be described. Engine 1 used here
As shown in FIG. 2, a six-cylinder gasoline engine has a fuel supply amount, an intake amount, an ignition timing, and the like, which are controlled by an engine controller 2. Engine controller 2
The main part is constituted by a microcomputer. Crank angle information is input from a crank angle sensor 4 provided opposite the crankshaft 3 of the engine 1. Further, the engine controller 2 drives a misfire indicator light 5 when misfire is determined. It is configured as follows.

The engine controller 2 obtains a unit crank angle signal (here, 5 ° crank angle information) and a reference position signal (here, 75 ° cylinder information before top dead center) from the crank angle sensor 4, and a built-in timer is provided. It is configured to emit a time pulse. The engine controller 2 includes a pre-elapsed time ratio calculation means 201 for calculating a pre _- elapsed time ratio T _en-1 / T _cn-1 and a post-elapsed time ratio calculation means 202 for calculating a post-elapsed time ratio T _en / T _cn. And the elapsed time ratio T _en
/ T before the elapse time ratio than _{cn T en-1 / T cn} -1 first by subtracting the
A first fluctuation value calculating means 203 for calculating a variation value [Delta] D _n,
A second fluctuation value calculating means 204 for calculating a second variation value [Delta] D n _{+ 1} according to the first variation value [Delta] D _n, the first change value [Delta] D _n from the second change value [Delta] D n _{+ 1} the subtraction to fluctuation width calculating a misfire index instruction value calculating means 205 for calculating on the basis of the first variation misfire degree instruction value by multiplying the value I _MF on the fluctuation range in the aforementioned equation (2)
When a function as the misfire judging means 206 for the misfire determination based on the misfire index instruction value I _MF (see Figure 3).

Here, the threshold value α for misfire determination used by the misfire determination means 20 is set based on experimental values. Here, an example of the characteristic is shown in FIG. Here, the threshold α has a tendency to decrease as the engine speed N increases.

FIG. 5 shows a misfire determination processing routine performed by the engine controller 2. Here, first, when the power is turned on, an initial value is set in each area, the timer is cleared, a circuit abnormality is determined, and the process proceeds to step a2. Here, the threshold α corresponding to the current engine speed N is calculated. Then, in step a3, a misfire determination is made based on the current misfire degree instruction value _IMF and the current threshold value α. If the current misfire degree instruction value _IMF is higher than the current threshold value α, the misfire indicator light 5 is switched to the driving state (indicated by a mark ○ in FIGS. 10 and 11).
Is switched off and returns.

If a 75 ° crank angle signal is interrupted during this misfire determination processing routine, the routine proceeds to a misfire degree instruction value _IMF calculation routine shown in FIG. Here, the crank angle θ
Is cleared, the value of the area Tn-1 is rewritten to the value of the area Tn, and the area Tn is rewritten to the value of the T timer. Further, the timers T and Tc are started. Subsequently, step b4
Then, the value of the area Ten-1 is rewritten with the value of the area Ten, and the area Ten is rewritten with the value of the timer Te. Thereafter, it waits for the crank angle θ to reach 5 °, and when it reaches, the step b6
Starts the timer Te.

In step b7, the area Tcn-1 is rewritten with the value of the area Tcn, and the value of the area Tcn is rewritten with the value of the timer Tc.

[0023] Revised variation value [Delta] D _n-1 After this change value [Delta] D _n, further, the current variation value _{ΔD n = (1 / T n} -1 2) × (T en /
_Tcn- Ten _-1 / _Tcn-1 ) is calculated. In step b10, the misfire degree instruction value I _MF = ΔD _n−1 × (ΔD _n−1 −Δ
D _n ) and return. The crank angle is 5 °
Every time, a crank angle calculation routine shown in FIG. 7 is executed.
Here, the value of the crank angle θ is added by 5 ° every 5 ° of the crank angle, and rewriting and return are performed.

FIGS. 8 to 11 show data obtained by such a misfire detecting device.

[0025] Here, in FIGS. 8 and 9 the engine speed N is shown the variation value [Delta] D _n at 1000rpm and 3000 rpm, the engine speed N in FIG. 10 and FIG. 11 10
00rpm and each misfire degree instruction value I _MF at 3000rpm are shown.

As is apparent from the above, at any engine speed N, the misfire degree instruction value _IMF is more than twice as large between the steady state and the misfire state, and the threshold value α is set relatively large. And detection accuracy can be improved. On the other hand, assuming that the variation value ΔD _n in FIGS. 8 and 9 is used as it is,
It is apparent that the threshold value α is set to a relatively small value, and the detection accuracy is low, especially at low rotations, between the steady state and the misfire, when the rotation speed is low.

[0027]

As described above, according to the present invention, the first variation value is obtained from the pre-elapsed time ratio and the post-elapsed time ratio, the second variation value is calculated according to the first variation value, and the first variation value is calculated. Then, the second fluctuation value is subtracted to calculate a fluctuation width, the fluctuation width is multiplied by the first fluctuation value to obtain a misfire degree indication value, and a misfire determination is made based on the misfire degree indication value. Thus, the misfire degree indication value more accurately includes misfire information, and the misfire detection accuracy can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a processing order of a misfire detection method according to the present invention.

FIG. 2 is a schematic diagram of an engine control system of an engine to which the misfire detection method of the present invention has been applied.

FIG. 3 is a functional block diagram of the engine controller of FIG. 2;

FIG. 4 is a characteristic diagram of a threshold value calculation map used by the engine controller of FIG. 2;

FIG. 5 is a flowchart of a misfire determination routine used by the engine controller of FIG. 2;

FIG. 6 is a flowchart of a misfire degree instruction value calculation routine used by the engine controller of FIG. 2;

FIG. 7 is a flowchart of a crank angle calculation routine used by the engine controller of FIG. 2;

8 is a diagram illustrating a variation value ΔD _n obtained from the engine of FIG.
It is a time-dependent characteristic diagram at 000 rpm.

FIG. 9 shows a variation value ΔD _n obtained from the engine of FIG.
It is a time-dependent characteristic diagram at 000 rpm.

10 is a misfire degree indication value I obtained from the engine of FIG. 2;
It is a time-dependent characteristic diagram of _MF at 1000 rpm.

11 is a misfire degree indication value I obtained from the engine of FIG. 2;
It is a time-dependent characteristic diagram of _MF at 3000 rpm.

FIG. 12 is a time-dependent characteristic diagram of a fluctuation value ΔD _n as a conventional example.

FIG. 13 is a waveform diagram of an engine crank angle.

[Explanation of symbols]

 1 engine 2 engine controller 3 crankshaft 4 crank angle sensor 5 misfire indicator α threshold

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Yasuhisa Yoshida 5-33-8 Shiba, Minato-ku, Tokyo / Inside Mitsubishi Motors Corporation (72) Inventor Mitsuhiko Yanagisawa 5-33-8 Shiba, Minato-ku, Tokyo No. Mitsubishi Motors Corporation

Claims (1)

(57) [Claims] 1. A first cylinder which is set before the top dead center of the first cylinder and which is the same as the first crank angle range and which is set to the second cylinder which reaches the next top dead center after the first cylinder. The respective elapsed time ratios with respect to the second crank angle region set before the first crank angle region are calculated as the previous elapsed time ratio, and the first cylinder setting of the second cylinder is set.
Each of the second crank angle range set before the first crank angle region and the third cylinder that follows the first crank angle region and reaches the top dead center next to the second cylinder and is the same as the first crank angle region. The elapsed time ratio is calculated as a post-elapsed time ratio, and then the previous elapsed time ratio is subtracted from the post-elapsed time ratio to calculate a first fluctuation value. Then, a second fluctuation value is calculated according to the first fluctuation value. A second variation value is subtracted from the first variation value to calculate a variation range, and the variation range is multiplied by the first variation value to obtain a misfire degree indication value. A misfire detection method characterized in that misfire is determined based on a value.