JPH0481548A - Combustion trouble detector for internal combustion engine - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to a device for detecting combustion abnormalities in an internal combustion engine, and particularly to a device for detecting misfires occurring in a multi-cylinder internal combustion engine based on rotational fluctuations of the engine. related to a device for

(Prior Art) Conventionally, microscopic fluctuations (ω fluctuations) in the engine rotational speed or the rotational angular velocity of the crankshaft accompanying ignition of a multi-cylinder internal combustion engine are monitored, and a certain cylinder is selected based on the amount of fluctuation. For example, Japanese Patent Application Laid-Open No. 61 (1982) discloses a device that detects combustion abnormalities that occur during combustion, especially so-called misfires in which the air-fuel mixture does not ignite, and notifies the driver of this fact and stops fuel supply to misfiring cylinders.
This method is known from Japanese Patent Application Laid-open No. 2-258955 and Japanese Unexamined Patent Publication No. 2-112646.

According to the device disclosed in Japanese Patent Application Laid-Open No. 61-258955, the amount of rotation fluctuation is determined from the deviation between the maximum rotation speed and the minimum rotation speed for each cycle of change in engine speed, and the amount of rotation fluctuation is determined from the previous rotation amount. It is determined that a misfire has occurred when the ratio of the current cycle value to the cycle value is smaller than a predetermined value.

Furthermore, according to the device disclosed in Japanese Unexamined Patent Publication No. 2-112646, the instantaneous rotation speed at a specific position between the top dead center signals is 1.
It is determined that a misfire has occurred when the rotational speed decreases by a predetermined value or more from the same instantaneous rotational speed before ignition.

(Problem to be Solved by the Invention) However, all of the above conventional devices detect misfires based on the amount of variation in engine rotational speed or crankshaft rotational angular velocity for each ignition cycle. A long-period (fuel cut) that does not involve fuel supply cutoff, such as a drop in engine speed due to
When there is a decrease in the rotational speed (compared to the TDC cycle), there is a risk of erroneously determining that a misfire has occurred.

Furthermore, when two cylinders that are adjacent in the ignition order misfire in succession, the conventional device has a problem in that the firing of the cylinder that is later in the ignition order cannot be detected.

Furthermore, in the conventional device, when the engine speed is high, the amount of decrease in the engine speed due to a misfire is not large, so there is a problem in that it is difficult to determine a misfire.

The present invention has been made in view of the above circumstances, and by detecting the rotational angular acceleration of the crankshaft and detecting the combustion abnormality based on the rotational angular acceleration, the combustion abnormality can be detected more accurately. An object of the present invention is to provide a combustion abnormality detection device for an internal combustion engine.

(Means for Solving the Problems) In order to achieve the above object, the present invention provides a combustion abnormality detection device that detects combustion abnormality in an internal combustion engine based on rotational fluctuations of the engine. crank pulse generating means for generating a pulse at a predetermined crank angle position according to the crank angle; and angular acceleration for detecting the amount of rotational angular acceleration of the crankshaft each time the pulse is generated based on the pulse generated by the crank pulse generating means. an average angular acceleration amount calculation means for calculating an average value of the angular acceleration amounts detected by the angular acceleration amount detection means; and an angular acceleration amount detection means detected by the angular acceleration amount detection means when the current pulse is generated by the crank pulse generation means. a combustion abnormality detection device for an internal combustion engine, comprising: a determination means for determining the occurrence of combustion abnormality in the engine based on a deviation between the angular acceleration amount and the average value calculated by the average angular acceleration amount calculation means. is provided.

(Function) Thus, based on the pulses generated at a predetermined crank angle position according to the rotation of the engine crankshaft, the amount of rotational angular acceleration of the crankshaft is detected each time the pulse occurs, and the amount of rotational angular acceleration is detected. An average value is calculated, and the occurrence of combustion abnormality is determined based on the deviation between the angular acceleration amount at the time of generation of the current pulse and the average value. This allows combustion abnormalities to be determined more accurately.

(Example) An embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is an overall configuration diagram of a fuel supply control device including a combustion abnormality detection device for an internal combustion engine according to the present invention. A throttle body 3 is provided in the middle, and a throttle valve 3' is arranged inside the throttle body 3. A throttle valve opening (θT11) sensor 4 is connected to the throttle valve 3', and outputs an electric signal according to the opening θTH of the throttle valve 3' and supplies it to an electronic control unit (hereinafter referred to as rEcUJ) 5. do.

A fuel injection valve 6 is provided for each cylinder between the engine 1 and the throttle valve 3' and slightly upstream of an intake valve (not shown) in the intake pipe 2, and each injection valve is connected to a fuel pump (not shown). It is electrically connected to the ECU 3, and the valve opening time for fuel injection is controlled by signals from the EC and U5.

On the other hand, an intake pipe absolute pressure (PB^) sensor 8 is provided immediately downstream of the throttle valve 3' via a pipe 7, and the absolute pressure PB^ detected by this absolute pressure sensor 8 and converted into an electric signal is detected by the intake pipe absolute pressure (PB^) sensor 8. is supplied to the ECU 3. In addition, an intake temperature (T^) sensor 9 is installed downstream of it.
EC by detecting the intake temperature T^ and outputting the corresponding electrical signal
Supply to U3.

An engine water temperature (Tw) sensor 10 attached to the main body of the engine 1 is composed of a thermistor, etc., and detects the engine water temperature (cooling water temperature) Tw and outputs a corresponding temperature signal to perform EC.
Supply to U3. Crank angle (CR) sensor 11, engine speed (Ne) sensor 12 and cylinder discrimination (CYL)
The sensor 13 is attached around the camshaft or crankshaft (not shown) of the engine 1. Crank angle sensor 1
1 outputs a signal pulse every time the crankshaft rotates by a predetermined angle (for example, 30 degrees), and the engine rotation speed sensor 12 outputs a signal pulse at a predetermined crank angle position every 180 degrees of rotation of the crankshaft of the engine L, that is, the intake air of each cylinder. Top dead center (TD) at the start of the stroke
Regarding C), a signal pulse (hereinafter referred to as rTDc signal pulse J) is output at a crank angle position before a predetermined crank angle, and the cylinder discrimination sensor 13 outputs a signal pulse (hereinafter referred to as rCYL signal pulse J) at a predetermined crank angle position of a specific cylinder. ), and each of these signal pulses is EC
Supplied to U3.

Further, a voltage sensor 14 that detects an output voltage VAcGF of a generator (not shown) connected to the engine 1 is connected to the ECU 3.
The electrical signal corresponding to the output voltage VACCF is supplied to the ECU 3. Furthermore, an air conditioner sensor 15 that detects the operating switch position of a near conditioner (not shown) mounted on the vehicle and a brake sensor 16 that detects whether the brakes of the vehicle are activated or not are connected to the ECU 3, and are connected to the ECU 3 to detect whether the brakes are activated or not. An electrical signal corresponding to the operation is supplied to the ECU 3.

Furthermore, the engine 1 is equipped with a traction control device (not shown) that reduces the amount of fuel supplied when detecting slip occurring between the drive wheels of the vehicle and the road surface to reduce engine output torque and eliminate slip. ) is provided, and a TC, S sensor 1 detects whether or not the fuel supply amount is reduced (leaning) by the device.
7 is connected to the ECU 3, and an electric signal corresponding to the detection is supplied to the ECU 3.

The ECU 3 includes an input circuit 5a having functions such as shaping input signal waveforms from various sensors, correcting voltage levels to predetermined levels, and converting analog signal values into digital signal values.
A central processing circuit (hereinafter referred to as "CPU") 5b that processes and executes a misfire detection program etc. to be described later, a storage means 5c that stores various arithmetic programs and arithmetic results etc. executed by the CPU 5b, and outputs misfire detection results, It is composed of an output circuit 5d that supplies a drive signal to the fuel injection valve 6, and the like.

The CPU 5b determines various engine operating states, such as a fuel cutoff operating state, based on various engine parameter signals from the above-mentioned sensors and other sensors (not shown), and performs fuel injection according to the engine operating state. The fuel injection time of the valve 6 is calculated, and a drive signal for opening the fuel injection valve 6 based on the calculated fuel injection time is supplied to the fuel injection valve 6 via the output circuit 5d.

The misfire detection procedure executed by the CPU 5b of the internal combustion engine fuel supply control device configured as described above will be described in detail below.

First, its outline will be explained using FIGS. 2 and 3.

FIG. 2 is a graph showing changes in the torque T that rotates the crankshaft each time each cylinder is ignited. When a misfire occurs in a certain cylinder, the torque T changes as shown by the broken line. Therefore,
It can be seen that misfire can be accurately detected by monitoring the change in torque T.

On the other hand, since the rotational angular acceleration of the crankshaft is generally proportional to the crankshaft torque T, a misfire can be accurately detected by monitoring the rotational angular acceleration.

The present invention focuses on this point, and measures the generation time interval of TDC signal pulses as shown in FIG. 3 under engine operating conditions suitable for detecting a misfire.
The time between the current pulse and the previous pulse of the signal pulse is t
n (The subscript n represents the current time, for example, n-1 is the previous time, n-
2 represents the time before the previous one), then the average angular velocity ω during this time is
n is the following formula (1) % Formula % Therefore, the angular acceleration αn is calculated based on the following formula (2).

In the present invention, this angular acceleration αn is monitored to detect a misfire. Specifically, the average value α^VE of this angular acceleration αn up to the previous time is calculated based on the following equation (3).

CtAVE=αn-4+ αn, ++an-2+ αn
-+ 00. (3) The deviation Δαn between the calculated average value α^VE and the current angular acceleration αn is calculated based on the following equation (4).

△αn=αn−α＾V E ”'... (
4) This calculated deviation Δαn has a negative value, and
When αn is smaller than a predetermined negative value (in other words, when the absolute value of the deviation △αn is larger than the absolute value of the predetermined value), the cylinder to be ignited between the current pulse and the previous pulse This is to determine that a misfire has occurred.

Next, a detailed description will be given of a procedure for detecting a misfire from a variation in the rotational angular acceleration of the crankshaft, which is executed by the CPU 5b based on the program flowchart shown in FIG. This program is executed every time a TDC signal pulse is input.

First, in step 101, it is determined whether the engine 1 is in a starting mode operation state. If this answer is affirmative (Yes), tr+phoN consists of a down counter that measures the elapsed time after the engine 1 leaves the starting mode operating state.
Set a predetermined time (for example, 10 seconds) to the IsT timer (Step +02), and then set the TDC that occurred between the next misfire and the next misfire.
A partial RAM of the storage means 5C, which stores the number of signal pulses in (mun+1) storage areas for each misfire, is initialized (step 103).

The detailed content of step 103 is shown in subroutine 5 in FIG.
Denoted by UB1. First, in step 201, the area number m is set to O, and the memory value MFro of the m-th storage area is set to O.
cm is set to 0 (step 202), and area number m is incremented (step 203). Step 20
4, the incremented area number m becomes the predetermined value mu
It is determined whether or not the area number m is larger than I (for example, 19), and if the answer is negative (No), the process returns to step 202. On the other hand, if the area number m exceeds the predetermined value mLMT, this subroutine SUB ends, Proceed to step 104 in FIG.

m L from number 0 by executing subroutine SUB 1
Memory values MFTDCO to MFT in the storage area up to MT number
All DCs 19 are set to O (initialized).

At step 104 in FIG. 4, flags F-cyLst and F-CYLI+2 representing each misfire of cylinders #1 to #4 are set.
゜F-cyu+3. F-CYL #4, continuous misfire count value 7ZIIFCN, occurrence count value of TDC signal pulses occurring during misfires that occur at certain intervals 7ZTD
CCNT, area number m, total value M F yoc of memory values MFTDCIII stored in the RAM of the storage means 5C
Set each sur+ to O and end this program.

On the other hand, if the answer to step 101 is negative (NO), tr+
It is determined whether the count value of the pr+oNIs timer is O (step 105). This answer is negative (No)
If so, the process proceeds to step 103, and if the answer to step 105 is affirmative (Yes), that is, a predetermined period of time ff1111ONIsT has elapsed after the engine l leaves the starting mode operating state, the process proceeds to steps 106 and 107.

In step 106, the engine cooling water temperature Tw is set to a predetermined value TW.
It is determined whether or not the temperature is higher than MFIIES (for example, 70°C), and in step 107, the intake temperature T^ is set to a predetermined value TAnr.
nes (for example, 50° C.).

If the answer to either step 106 or step 107 is negative (NO), it is determined that the engine operating state is not suitable for detecting a misfire, and the process proceeds to step 103. On the other hand, if the answers to both steps 106 and 107 are affirmative (Yes), the process proceeds to step 108.

In step 108, it is determined whether the engine 1 is currently in a fuel cut operation state. This answer is affirmative (Ye
s), then the tFCIIFDLY timer is set to a predetermined time tpcr+p, which consists of a down counter that measures the elapsed time after the engine 1 leaves the fuel cut operation state.
oLy (for example, 5 seconds) (step 109).
) The process proceeds to step 103, and if the answer to step 108 is negative (No), it is determined whether the count value of the FC1IFDLY timer is 0 (step 110). If the answer is negative (No), the process proceeds to step 103, while if it is affirmative (Yes), that is, the fuel cut role predetermined time t
If FCIIFDLY has elapsed, step 111
Proceed to ~114.

In step 111, it is determined whether the absolute value of the degree of change ΔOn+ of the throttle valve opening degree en+ is smaller than a predetermined value (3TIu+pppt (for example, 2 degrees)), and in step 112
Then, it is determined whether the absolute value of the change degree △PB of the intake pipe internal shoe pressure PB^ is smaller than a predetermined value PB gate FDIl:L (for example, 40 mm 1 g), and in step +13, it is determined whether the vehicle is in a cruising state. Determine whether The step +
13 is determined based on vehicle running speed and throttle valve opening OT1.
This is done by determining whether or not the intake pipe absolute pressure PBΔ and the intake pipe absolute pressure PBΔ are respectively within predetermined value ranges. In step 114, it is determined whether a flag jvAg+ (described later with reference to FIG. 6), which is set to 1 when there is a disturbance that tends to cause a misfire, is O.

If any of the answers to Steps 111 to 114 is negative (NO), it is determined that the engine operating state is not suitable for detecting a misfire, and the process proceeds to Step 103; positive(
If yes), the process advances to step 115.

The flag F VARI determined in step 114 is set in subroutine 5UB6 shown in FIG. This subroutine 5UB6 is executed separately from the execution of the program shown in FIG. 4 every time an FDC signal pulse is input.

In step 301 of FIG. 6, it is determined whether or not there has been a sudden change in the engine speed Ne. Specifically, voltage sensor 1
The absolute value of the deviation between the previous value and the current value of the generator output voltage VACGF detected by 4 is the predetermined value △■＾CGFL
By determining whether or not it is larger than r17, it is determined whether or not there is a sudden change in the engine speed Ne. Step 302
Then, based on the output of the air conditioner sensor 15, it is determined whether the near conditioner operation switch has been switched from OFF to ON or from ON to OFF between the last execution of this program and the current execution of this program. .

In step 303, based on the output of the brake sensor 16, it is determined whether or not the brake has transitioned from activation to non-activation or from non-activation to activation between the previous execution of this program and the current execution of this program. Step 304
Then, based on the output of the TCS sensor 17, it is determined whether or not the supplied air-fuel mixture is lean (misfire is likely to occur).

If any of the answers from step 30+ to step 304 is affirmative (Yes), it is assumed that there is a disturbance that is likely to induce a misfire, and step 305 is executed. At step 307, the flags F to VARI are set to 1, and this subroutine 5U
End B6. In step 305, a predetermined time jvARI is set in a jvAR+timer consisting of a down counter that measures the elapsed time after the disturbance ceases to exist.

On the other hand, the answers to steps 301 to 304 are all negative (
No) If so, then VAT! It is determined whether the count I value of the I timer is 0 (step 306). If this answer is negative (No), proceed to step 307, while affirmative (Y)
es), that is, a predetermined time jv after the disturbance ceases to exist.
When ^R■ has elapsed, the flag F VARI is set to O and this subroutine 5UB6 is ended.

Returning to step 115 in FIG. 4, when a CYL signal pulse is generated between the previous pulse generation and the current pulse generation of the TDC signal pulse by executing another subroutine,
The flag F-cy is set to 0 when it does not occur.
+, rLs is 1, but it is determined whether or not. If the answer is affirmative (Yes), the cylinder specific count value 7ZTDC is set to O (step 116); on the other hand, if the answer to step 115 is negative (No), step 116 is skipped and the process proceeds to step 117.

Stellaf 11 foot count value r2. Increment the values of TDCCNT and 7ZTDC. Therefore, the CYL signal pulse and the TDC signal pulse have a generation timing relationship as shown in FIG. 3, and according to the generation of the TDC signal pulse, #3. #4. #2゜When the #1 cylinder ignites, the correspondence relationship between the value of the count value 72 DC and the number (CYL#) of the cylinder that should be ignited when the TDC signal pulse corresponding to this value is generated. is as shown in FIG. 7(a). FIG. 7(a) shows the changes in each value of the TDC signal (each time the pulse generation water program is executed).

In step 118, the value 7Z, DCCN, which was incremented in step +17 is changed to the overflow value (
For example, 2047) is determined. If the answer is affirmative (Yes), it is determined that there is no abnormal misfire condition and the same process as in step 103 is executed (step 119), whereas the answer to step 118 is negative (NO).
If so, skip step 119 and proceed to step 120.

In step 120, the generation time interval of the TDC signal pulse is measured (calculated), and the rotational angular acceleration of the crankshaft is calculated based on the measured time, and a misfire is determined from the calculated angular acceleration. This will be explained with reference to subroutine 5UB3 shown in FIG.

First, in step 401, the formulas (1) and (2)
.

Based on (3), the average angular velocity ωn, the angular acceleration αn, and the average value α^VE are calculated.

The time un between the current pulse and the previous pulse of the TDC signal pulse is determined by counting the clock pulses generated at a certain time interval for each generation interval of the signal pulse output from the crank angle sensor 11, and calculating the count value. It is obtained by adding the TDC signal pulses between the current pulse generation time and the previous pulse generation time.

Furthermore, in step 402, the deviation △α is calculated based on the formula (4).
n is calculated, and it is determined whether the calculated deviation Δαn is smaller than a predetermined negative value M F DELLI-IT (step 403). The predetermined value M F DELLMT is a negative value determined according to the engine speed Ne and the intake pipe absolute pressure PB^ representing the engine load, and is
The higher e or the greater the absolute pressure PB^ (higher load), the smaller the value (the larger the absolute value of M F DIELLllT)
It is set to . Specifically, the predetermined value MFDEL
LlIT is obtained from the table shown in FIG. 9, and when the intake pipe absolute pressure PB^ is the value PB2, when the engine speed Ne is less than the value Ne1, the value M F oEu,
r+T+, and the engine speed Ne becomes the value Ne7(Ne
7) Above Ne+), the value M F DELLIIT? (M
FDELL gate T7<MF DELIJITI), and when the engine speed Ne is between the value Ne+ and Ne7, the value M F DEL increases as the engine speed Ne increases, and from 11TI, the value M F DELLMT? gradually decrease to When the intake pipe absolute pressure PBA is the value PBI (PBI<PB2), the predetermined value M F DELLI1 is set to a larger value than when the absolute pressure Ps^ is the value PR2, and the absolute pressure P
When B^ is between the value PB2 and PBI, the predetermined value MF
DELLIIT is determined by interpolation subtraction.

If the answer to step 403 in FIG. 8 is affirmative (Yes), it is determined that a misfire has occurred in the cylinder to be ignited according to the previous TDC signal pulse, and the continuous misfire count value 7?1FCNT is incremented (step 404). , see FIG. 7(a)) This subroutine 5UB3 is terminated, and at the same time, the main program of FIG. 4 is terminated. In Figure 7(a), “
The X mark in the "Misfire Detection" column indicates that a misfire was detected during the execution of the program shown in FIG. 4 due to the generation of the TDC signal pulse this time. That is, this indicates that there was a misfire in the cylinder that should have been ignited according to the previous TDC signal pulse (the cylinder in parentheses in the "Misfire Detection" column).

On the other hand, if the answer to step 403 is negative (NO), the consecutive misfire count value 7ZnpcNT is greater than O, and the predetermined value M
It is determined whether or not it is smaller than FCNTLIIT (step 405). The predetermined value M FcNn,hr is a value corresponding to the number of cylinders, and is 4 in this embodiment. Step 405
The answer is negative (No), that is, step 403 was negative during the previous program execution and no misfire was detected (7ZllFCNT = O), or all cylinders misfired consecutively (ntycNt = 4). If so, set the consecutive misfire count value 7Zr+FcNr to 0 (step 4).
06) At the same time as finishing this subroutine 5UB3,
Exit the main program shown in the figure. In addition, when all cylinders misfire continuously, it is a natural misfire that occurs when combustion is not normally occurring, for example, when fuel is not being supplied due to a fuel cut during deceleration, etc. ends this subroutine 5UB3 without incrementing the continuous misfire count value 7Zmpc+n.

On the other hand, if the answer to step 405 is affirmative (Yes), the process advances to steps 407 and 408 to identify the cylinder in which the misfire detected previously occurred (step 407), and to determine whether the misfire rate has exceeded a predetermined level. A determination is made (step 408).

The detailed contents of step 407 will be explained with reference to subroutine 5UB2 in FIG. First, in step 501, the misfire cylinder specific value 7Z11FCYL is set to the same value as the number of consecutive misfires counted up to the previous time, 72MFCN. Next, in steps 502, 504, 506, and 508, it is determined whether the cylinder specific count value is 72TDC, I, or 2.3.4, and if it is 1, the flag F-CYL is set.
# Set 2 to 1, similarly if it is 2 then F-cYt**, if it is 3 then F-cYLx3, if it is 4 then F-cy+, *
4 to 1, respectively (steps 503, 505, 50
7.509). That is, cylinder specific count value 72. Based on the TDC value, the cylinder detected last time or in which a misfire occurred (the cylinder to be ignited according to the TDC signal pulse before the previous time) is specified (see FIG. 7(a)).

Next, in steps 510 to 5]7, the misfire is 1'DC.
Each cylinder in which misfires occur continuously is identified every time a signal pulse is input. In step 510, misfiring cylinder specific value 7
Zr+FcyL is decremented, and it is determined whether the decremented misfiring cylinder specific value 7ZhpcyL is O (step 511). If this answer is negative (No), that is, the misfire occurs continuously every time the TDC signal pulse is input, and not only the last time <71r+FcNr=1) but also the time before the previous time (7ZMFCNT・2) or even the time before that ( 7
211FCNT=3) When a misfire is detected, the cylinder specific count value 7ZTDC is decremented (
Step 5] 2) Determine whether or not the decremented 7ZTDC is O (step 513). If the answer to this question is affirmative (Yes), 7ZTDC should be set to 4 since 72rpc has the property of cyclically taking values from 1 to 4 (step 514); on the other hand, if the answer to step 513 is negative (no), step 514) Skip 514 and return to step 502. That is, steps 512 to 51
4 decrements the cylinder identification count value 7;l, TDC in preparation for the re-execution of the misfiring cylinder identification steps 502 to 509, and executes the steps 502 to 509 using this decremented 7ZTDC, so that the misfire cylinder identification count value 7; This is intended to identify the cylinder in which the previously detected misfire occurred.

As a result of the above execution, if step 511 becomes affirmative (Yes), the cylinder specific count value 72TDC is set in step 51.
Step 515 to return to the value before decrementing step 2
to 517 are executed. That is, steps 512 to 514
The value obtained by adding the consecutive misfire count value 72MFCNT counted up to the previous time to the 7hoc after execution and further subtracting 1 is set as 7ZTDC (Step 515), and this 7ZTDC
Only when is over 4, subtract 4 and set the value as 72rpc (Step 5] 6.5] 7), this subroutine 5
UB2 is ended and the process proceeds to step 408 in FIG.

Next, the detailed contents of step 408 will be explained with reference to subroutine 5UB4 in FIG. 11. First step 6
At 01, the storage means 5 stores the TDC signal pulse value 'r1tonoccNT counted up to the previous time that occurred during the misfire.
The stored value M F r is stored in the mth area of the RAM of c.
oam (subscript m is O based on the next step 602)
to a predetermined value mLjT.

(See Figure 7(a)).

Next, increment the area number m in step 602.
) and set the value 72yroccNr to O (step 603). It is determined whether or not m incremented in step 602 is greater than a predetermined value mLIIT (step 604). If the answer is negative (No), this subroutine 5UB4 is ended and the process proceeds to step 406 in FIG. 8, where the count value is When rlrIFcNT is set to O and subroutine 5UB3 is ended, the main program shown in FIG. 4 is also ended. Therefore, by executing steps 601 to 604 in FIG. 11, M
FTDCO=2, MFTDC+=4.

MFTDC2=8. ...-MFTDc+s=7. M
FTDC+9=5 is set.

On the other hand, if step 604 becomes affirmative (Yes) (when m=19 in FIG. 7(a) is incremented in step 602), steps 605 to 611 are executed, and an abnormality determination based on the misfire rate is performed. That is, m is decremented (step 6o5) and control variable k is set to O (step 6o6). Next, the RAM of the storage means 5c
Add the stored value MFrpck (the subscript takes a value from O to m1JIT in the next step 608) stored in the number area of 2 to the total value M F TDC6U+1 to create a new total value M F rpcsur+ Step 607
) Increment k (step 6o8).

The combination value M F is the sum of the number of TDC signal pulses between misfires and therefore roughly represents the reciprocal of the misfire rate.

In the next step 609, the sum value M F TDC3 [Jll calculated in step 607 is the reference value SUMLIIT
(for example, 2000). If the answer is affirmative (Yes), it is determined that there is no abnormal misfire condition and the process proceeds to step 612; if the answer is negative (NO), k incremented in step 608 is set to the predetermined value mLMT.
It is determined whether or not the value is larger than (Step 6] O).

If the answer is negative (No), the process returns to step 607, while if affirmative (Yes), the process proceeds to step 611 and step 612, in which the flag I'-hpNc is set to 1, indicating that an abnormal misfire is occurring. Based on the setting of the flag, ECIJ
5 takes fail-safe measures and also issues a warning to the driver.

FIG. 7(b) shows an example value of the total value MFDC8LIM calculated by executing steps 605 to 611 above according to the control variable k. In other words, during the 20 misfires (consecutive misfires are counted as 1), the total number of TDC signal pulses that occurred between misfires M F Toc
If sur+ does not reach (for example, 2000), the misfire frequency is too high and it is determined that the misfire is abnormal.

Next, by executing step 612, FIG.
), the stored values M F rpcm stored in the RAM of the storage means 5c are transferred to the side of the smallest number m-+ of the area number m, and step 601.6
Prepare for the next execution on 07.609.

The detailed contents of step 612 will be explained with reference to subroutine 5UB5 shown in FIG.

First, in step 701, the control variable k is set to 0, ++ + is transferred to the stored value M P rpc+ to MFrock (step 702), and k is incremented (step 7o3).

The incremented k in the next step 704 is (n+
un-+). If the answer is negative (No), the process returns to step 702;
s), then k is incremented in step 703.
The stored value MFrock (MFDC+9) is set to 0 (step 7o5), and the total value MFTDC3U11 is also set to O (step 706), where the area number is MFrock (=19 in this embodiment).
At the same time, 5UB4 in FIG. 11 is ended and the process proceeds to step 40G in FIG. 8. After executing step 406, the subroutine 5UB3 is terminated, and at the same time, the main program shown in FIG. 4 is also terminated.

Since the area number m is rrB7H (19 in this embodiment) when the main program shown in FIG. 4 is executed next time (see FIG. 7(d)), steps 601 to 604 in FIG. If so, steps 605 to 612 will be executed at that execution time. That is, the entire memory value MFrocm is not updated, but only one (mu+r++) memory content is updated to a new one by subroutine 5UB5 in FIG.
It is determined whether or not there is an abnormal misfire condition based on the updated total value M F rocsuIX.

Although the above embodiments have been described using a four-cylinder engine, the invention is not limited to this, and the present invention can be applied to multiple-cylinder engines including single-cylinder engines.

(Effects of the Invention) As described in detail above, the present invention provides a combustion abnormality detection device that detects combustion abnormality in an internal combustion engine based on rotational fluctuations of the engine. crank pulse generating means for generating a pulse at a position; angular acceleration amount detecting means for detecting an amount of rotational angular acceleration of the crankshaft each time the pulse is generated based on the pulse generated by the crank pulse generating means; an average angular acceleration amount calculation means for calculating an average value of the angular acceleration amounts detected by the acceleration amount detection means; and an angular acceleration amount detected by the angular acceleration amount detection means at the time of the current pulse generation of the crank pulse generation means and the average value. Since the apparatus includes a determining means for determining the occurrence of combustion abnormality in the engine based on the deviation from the average value calculated by the angular acceleration amount calculating means, the combustion abnormality can be detected more accurately.

In particular, since the rotational angular acceleration of the crankshaft is detected and a misfire is detected based on the angular acceleration, the misfire detection accuracy in a high engine rotation range is improved compared to a conventional device that detects a misfire based on the rotational angular velocity.

That is, as shown in Fig. 13, the difference between the angular acceleration α during normal operation and that during a misfire increases as the engine speed Ne increases (and as the engine load increases), so the difference becomes larger in the high engine speed range. The misfire detection accuracy is improved compared to the conventional device based on the rotational angular velocity, which is a type in which the misfire is not detected.

In addition, even in cases where conventional devices would determine that a misfire has occurred because the engine speed fluctuates even though there is no misfire, such as when upshifting in an AT car, the angular acceleration Since this does not change, it is possible to avoid misjudgment as a misfire.

In addition, continuous misfires can be reliably detected because the judgment is made by comparing with the average value.

[Brief explanation of drawings]

Fig. 1 is an overall configuration diagram of a fuel supply control device including a combustion abnormality detection device for an internal combustion engine according to the present invention, Fig. 2 is a time change graph of crankshaft torque 1゛, and Fig. 3 is a CYL
Timing chart of signal pulse, TDC signal pulse,
FIG. 4 is a flowchart of the misfire detection main program executed by the CPIJ5b in FIG.
, 11.12 Figures show subroutines 5UBI and 5UB6, respectively.
, 5UB3゜5UB2, 5UB4, 5tJB5, FIG. 7 is a timing chart showing the time changes of each value determined by the execution of the main program and each subroutine, and FIG. 9 is a predetermined flowchart used in step 403 of FIG. The table of values M F DELLIIT, FIG. 13, is a graph of changes in angular acceleration α with respect to engine speed Ne. DESCRIPTION OF SYMBOLS 1... Internal combustion engine, 5... Electronic control unit (ECU), 8... Absolute pressure sensor, 12... Engine rotation speed sensor.

Claims (3)

[Claims] 1. A combustion abnormality detection device for detecting combustion abnormality in an internal combustion engine based on rotational fluctuations of the engine, comprising a crank pulse generating means for generating a pulse at a predetermined crank angle position according to rotation of a crankshaft of the engine; angular acceleration amount detection means for detecting the amount of rotational angular acceleration of the crankshaft each time the pulse is generated based on the pulse generated by the generation means; and an average value of the angular acceleration amount detected by the angular acceleration amount detection means. Based on the average angular acceleration amount calculation means to be calculated and the deviation between the angular acceleration amount detected by the angular acceleration amount detection means at the time of the current pulse generation of the crank pulse generation means and the average value calculated by the average angular acceleration amount calculation means. A combustion abnormality detecting device for an internal combustion engine, comprising: a determining means for determining the occurrence of a combustion abnormality in the engine. 2. 2. The combustion abnormality detection device for an internal combustion engine according to claim 1, wherein the determining means determines that a combustion abnormality has occurred in the engine when the deviation is larger than a predetermined value. 3. 3. The combustion abnormality detection device for an internal combustion engine according to claim 2, wherein the predetermined value is a value determined according to engine load and engine rotation speed.