JP2694764B2 - Misfire detection device for internal combustion engine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a misfire detection device for an internal combustion engine, and more particularly to a misfire detection device for an internal combustion engine that detects a misfiring cylinder of a multi-cylinder internal combustion engine.

[0002]

2. Description of the Related Art An internal combustion engine which detects a rotation angle of a crankshaft at every predetermined angle, detects a fluctuation (rotation fluctuation) of a detection cycle of the rotation angle, and detects a misfire cylinder based on the rotation fluctuation of the crankshaft. 2. Description of the Related Art An engine misfire detection device is conventionally known as a known technology (for example, see Japanese Unexamined Patent Application Publication No.
No. 49955).

[0003] However, in the misfire detection device, when the engine speed is low, the so-called "swinging" of the engine occurs, so that the rotational fluctuation amount of the crankshaft is large,
There is a disadvantage that low frequency noise is mixed in the detection of the rotation fluctuation amount and it is difficult to perform accurate misfire detection.
On the other hand, when the engine speed is high, high frequency noise is apt to be mixed in when detecting the fluctuation amount of the crankshaft because of the torsional vibration of the crankshaft and the vibration caused by the so-called "chatter" of the journal. There was a drawback that it was difficult to perform.

Accordingly, the applicant of the present application has already proposed a misfire detection apparatus which improves the accuracy of misfire detection by providing filter means and removing noise components by the filter means (December 4, 1991). Patent application submitted on the day).

In the misfire detecting device, a high-pass filter and a low-pass filter are provided as filter means, and the high-pass filter is used when the engine speed is low, and the low-pass filter is used when the engine speed is high. The noise component is removed to improve the accuracy of misfire detection.

[0006]

However, in the above-mentioned misfire detection device, a predetermined judgment value for misfire judgment is set without considering the type of transmission, the engagement state of the clutch, and the like. There was such a problem.

That is, as is well known, the automatic transmission mechanism (A
T) has a built-in torque converter for shifting the engine torque, etc., and is therefore a manual transmission mechanism (MT).
The rotation fluctuation is less likely to occur, and the rotation fluctuation amount at the time of misfire is smaller than that of the manual transmission mechanism. Therefore, AT
When using the same predetermined judgment value in use and the use MT, the transmission mechanism is set larger than the desired value is predetermined determination value when AT, misfire even though the engine cylinders is misfiring
And there is a possibility that an erroneous determination that no, whereas transmission MT
If, the predetermined judgment value is set smaller than the desired value, and the engine cylinder is not misfiring, it is misfiring.
Then, there is a problem that an erroneous determination may be made.

Also, regarding the engaged state of the clutch, fluctuations in the rotation of the crankshaft are less likely to occur when the clutch is engaged than when the clutch is disengaged. Therefore, when the same predetermined determination value is used regardless of the clutch engagement state, for example, when the clutch is in the engagement state, the predetermined determination value is set smaller than the desired value, and the engine cylinder does not misfire. However, there is a risk of misjudging that the engine has misfired.On the other hand, when the clutch is in the disengaged state, the predetermined judgment value is set larger than the desired value, and the engine cylinder is misfiring. There is a risk that it may be erroneously determined that a misfire has not occurred, and it is difficult to perform accurate misfire detection.

Further, in the above-mentioned misfire detection device, since the fluctuation amount (absolute value) of the crankshaft rotation becomes small when the filter is used, the judgment value is obtained by searching different maps when the filter is used and when the filter is not used. There is a problem that the control system is complicated because the misfire is detected by setting it.

Further, in the above-mentioned misfire detecting device, the noise component is removed by using the high-pass filter and the low-pass filter as described above, but the high-pass fill and the low-pass filter discriminate the misfiring cylinder position (T).
There is a problem that the DC position) is different. That is,
18 when the low pass filter is used compared to when the filter is not used
The misfiring cylinder is discriminated with a phase delay of 0 °, and the misfiring cylinder is discriminated with a phase delay of 360 ° when the high-pass filter is not used, as compared to when the filter is not used. There is a problem that misidentification of cylinders may occur.

The present invention has been made in view of the above problems, and can accurately determine the misfiring cylinder regardless of the type of transmission, the engaged state of the clutch, and the used state of the filter. An object of the present invention is to provide a misfire detection device for an internal combustion engine.

[0012]

In order to achieve the above object, the present invention provides a crank angle detecting means for detecting a rotation angle of a crank shaft at every predetermined angle, and a cycle of a detection signal obtained by the crank angle detecting means. Misfire detection of an internal combustion engine including cycle fluctuation detection means for detecting fluctuations, and misfire determination means for comparing the detection result of the cycle fluctuation detection means with a predetermined judgment value to judge whether the internal combustion engine has misfired or not. The device has a speed switching mechanism of either one of an automatic transmission mechanism and a manual transmission mechanism, and a determination value switching means for switching the predetermined determination value according to the type of the speed switching mechanism. I am trying.

The present invention also detects the engine speed.
Rotation speed detection means and the rotation speed detection means
Outputs only the detection signal of a specific frequency band among the detection signals
A plurality of filter means, and switching the judgment value
The predetermined judgment value selected by the filter means of the filter means.
Equipped with correction means to correct the weight loss according to the output frequency band
It is characterized by having.

Further, according to the present invention, a crank angle detecting means for detecting the rotation angle of the crank shaft for each predetermined rotation angle, and a rotation speed detecting means for detecting the engine speed based on the detection result of the crank angle detecting means. , It is determined whether or not the internal combustion engine has misfired by comparing the cycle fluctuation detection means for detecting the cycle fluctuation of the detection signal obtained by the crank angle detection means with the detection result of the cycle fluctuation detection means with a predetermined judgment value. In a misfire detection device for an internal combustion engine, comprising: a misfire determination means for operating a speed change mechanism capable of speed change in a plurality of stages, and engaging a clutch provided between the speed change mechanism and the engine. It has a judgment value switching means for switching the predetermined judgment value in accordance with it, and outputs only a detection signal of a specific frequency band among the detection signals obtained by the rotation speed detection means. A plurality of filter means, is characterized in that the predetermined decision value selected by said predetermined value switching means and a correction means for decreasing correction according to the output frequency band of the filter means.

Further, the present invention further comprises a deciding means for deciding whether or not the filter means should be used according to the engine speed detected by the engine speed detecting means, and based on a result of the deciding means, When a misfire is detected by the misfire determination means while the filter means is being used, a misfire cylinder determination means for sequentially determining the misfire of the cylinders having the same time delay with respect to the misfire detection timing is provided. Is characterized by.

Further, the filter means includes at least a high-pass filter through which frequency components above a first predetermined frequency band pass and a low-pass filter through which frequency components below a second predetermined frequency band pass.

[0017]

According to the above construction, the predetermined judgment value can be set to the desired value by the judgment value switching means according to the type of the speed switching mechanism. Further, it becomes possible to set the predetermined judgment value to a desired value according to the engagement state of the clutch by the judgment value switching means.

Further, by reducing the predetermined judgment value selected by the judgment value switching means in accordance with the output frequency band of the filter, the predetermined judgment value is set to a value such that misfire misjudgment does not occur.

Further, when the engine misfire is detected when the filter is used, the misfire of the cylinder having the same time delay can be detected at the same misfire detection timing.

Specifically, when a high-pass filter and a low-pass filter are provided as the filter means, a predetermined judgment value can be set according to the output frequency band of the high-pass filter or the low-pass filter. Whichever is used, the misfire of the same specific cylinder can be determined at the same misfire detection timing.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is an overall configuration diagram showing an embodiment of a misfire detection device for an internal combustion engine according to the present invention.

In the figure, reference numeral 1 denotes a DOHC in-line 4-cylinder internal combustion engine (hereinafter simply referred to as "engine") in which each cylinder is provided with an intake valve and an exhaust valve (not shown). A throttle body 3 is provided in the middle of an intake pipe 2 of the engine 1, and a throttle valve 3'is arranged inside thereof. A throttle valve opening (θTH) sensor 4 is connected to the throttle valve 3 ′, and outputs an electric signal according to the opening of the throttle valve 3 ′ to output an electronic control unit (hereinafter referred to as “ECU”) 5. To supply.

The fuel injection valve 6 is provided for each cylinder in the middle of the intake pipe 2 and slightly upstream of the intake valve between the engine 1 and the throttle valve 3 ', and is connected to a fuel pump (not shown). It is electrically connected to the ECU 5, and the EC
The valve opening time of fuel injection is controlled by a signal from U5.

A branch pipe 7 is provided downstream of the throttle valve 3'of the intake pipe 2, and an absolute pressure (PBA) sensor 8 is attached to the tip of the branch pipe 7. The PBA sensor 8 is electrically connected to the ECU 5, and
The absolute pressure PBA is converted into an electric signal by the PBA sensor 8 and supplied to the ECU 5.

An intake air temperature (TA) sensor 9 is mounted on the wall of the intake pipe 2 downstream of the branch pipe 7, and the intake air temperature TA detected by the TA sensor 9 is converted into an electric signal.
It is supplied to the ECU 5.

An engine water temperature (TW) sensor 10 composed of a thermistor or the like is mounted on the cylinder peripheral wall filled with cooling water of the cylinder block of the engine 1, and the engine cooling water temperature TW detected by the TW sensor 10 is converted into an electric signal. And is supplied to the ECU 5.

Cylinder discrimination (CYL) is performed at a predetermined position around the cam shaft or crank shaft (not shown) of the engine 1.
A sensor 11 and a crank angle (CRK) sensor 12 are respectively mounted.

The CYL sensor 11 outputs a pulse signal (hereinafter referred to as "CYL signal pulse") at a predetermined crank angle position of a specific cylinder for every two rotations of the crankshaft, and the C signal is output.
The YL signal pulse is supplied to the ECU 5.

The CRK sensor 12 has a pulse signal (hereinafter referred to as "CR" at a predetermined crank angle cycle (for example, 30 ° cycle)).
And outputs the CRK signal pulse to the ECU 5.

The spark plug 13 of each cylinder of the engine 1 is
It is electrically connected to the ECU 5, and the ignition timing is controlled by the ECU 5.

A wide-range oxygen concentration sensor (hereinafter referred to as "LAF sensor") 16 is provided in the exhaust pipe 15 of the engine 1, and the oxygen concentration in the exhaust gas detected by the LAF sensor 16 is provided. Is converted into an electric signal and supplied to the ECU 5.

The transmission 17 is a known automatic transmission mechanism (A
T) or a manual transmission mechanism (MT), which is interposed between wheels (not shown) and the engine 1 and has an ECU 5
Is electrically connected to

A drive wheel speed (VW) sensor 18 and a driven wheel speed sensor 19 (vehicle speed (VSP) sensor) for detecting the rotational speeds of drive wheels and driven wheels (not shown) are connected to the ECU 5, respectively. , The driving wheel speed VW detected by the VW sensor 18 and the vehicle speed VSP detected by the VSP sensor 19 are respectively converted into electric signals to EC.
It is supplied to U5.

Various switches such as an air conditioner switch 20, a brake light switch 21, and a power steering switch 22 are connected to the ECU 5, and the ON / OFF signals of these various switches are transmitted to the ECU 5.
Supplied to

Further, other electric devices 23 such as headlights and room fans are connected to the ECU 5, and ON / OFF signals from these electric devices 23 are supplied to the ECU 5.

Therefore, the ECU 5 forms an input signal waveform from the above various sensors, corrects the voltage level to a predetermined level, and converts the analog signal value into a digital signal value. A central processing circuit (hereinafter referred to as "CPU") 5b, a storage means 5c including a ROM and a RAM for storing various calculation programs executed by the CPU 5b, various maps to be described later, calculation results, and the like; An output circuit 5d for supplying a drive signal to the valve 6 and the spark plug 13 is provided.

Further, the CPU 5b determines an engine operating state such as a feedback control operating region or an open loop control operating region according to the oxygen concentration in the exhaust gas based on the above-mentioned various engine parameter signals, and according to the engine operating state, The fuel injection time TOUT of the fuel injection valve 6 is calculated based on the equation (1), and the result is stored in the storage means 5c.
(RAM).

TOUT = Ti × KLAF × KLS × K1 + K2 (1) Here, Ti is the engine speed NE and the intake pipe absolute pressure P.
A basic fuel injection time set according to BA,
As a Ti map for determining the Ti value, a TiCR map for the starting mode and a TiM map for the basic mode are stored in the storage means 5c (RAM).

KLAF is an air-fuel ratio correction coefficient and is set so that the air-fuel ratio detected by the LAF sensor 16 matches the target air-fuel ratio during the air-fuel ratio feedback control.
During the open loop control, it is set to a predetermined value according to the engine operating state.

KLS is a lean correction coefficient, which is set to a predetermined value according to the operating region by the lean coefficient immediately before the fuel cut (fuel supply stop). In the start mode, the KLS value is always set to "1.0".

K1 and K2 are correction coefficients and correction variables calculated according to various engine parameter signals, and optimization of various characteristics such as fuel consumption characteristics and acceleration characteristics according to the engine operating condition for each cylinder. It is set to a predetermined value as shown.

Further, the CPU 5b calculates the slip ratio λ according to the driving wheel speed VW18 and the vehicle speed VSP based on the equation (2).

[0044]

(Equation 1) That is, the CPU 5b controls the drive wheel slip control system (traction control system, hereinafter “TCS”).
), And reduces the torque of the drive wheels in accordance with the slip ratio λ to suppress excessive slip of the drive wheels.

FIG. 2 shows a CYL signal pulse output from the CYL sensor 11 and a C output from the CRK sensor 12.
6 is a time chart showing the generation timing of an RK signal pulse.

The CRK signal pulses generate, for example, 24 signal pulses at equal intervals while the crankshaft of the engine 1 makes two revolutions, that is, clock pulses at crank angle intervals of 30 °, for example. The ECU 5 counts a predetermined number of clock pulses of the CRK signal pulse, generates a TDC determination signal every six pulses (every 180 ° rotation of the crankshaft), and detects a reference crank angle position of each cylinder. Further, the ECU 5 determines whether the CRK signal pulse generation time interval CR
The ME is measured, and these CRME values are added over the time interval of generation of the TDC determination signal to calculate the ME value.
The engine speed NE, which is the reciprocal of the value, is detected.

The CYL signal pulse is a predetermined crank angle position (for example, 90 °) before the TDC discrimination signal generating position indicating the end of the compression stroke of a specific cylinder (for example, # 1CYL).
° BTDC) and the T immediately after the generation of the CYL determination signal.
A specific cylinder number (for example, # 1CY
L) is set.

Further, the ECU 5 sends the TDC discrimination signal, C
A crank angle stage (hereinafter, referred to as “stage”) from a reference crank angle position of each cylinder is detected based on the RK signal pulse. That is, when the CRK signal pulse C1 detected at the same time as the generation of the TDC discrimination signal is generated at the TDC position at the end of the compression stroke of # 1CYL, EC
U5 is # 0 of # 1CYL by the CRK signal pulse C1.
, # 2 stage,..., # 2
Three stages are sequentially detected.

Therefore, the CPU 5b has a misfire detecting means for detecting the misfire of the engine cylinder, and its method will be described in detail below.

FIG. 3 is a flow chart showing the overall construction of the misfire detecting means. This program is executed in synchronization with the generation of the TDC discrimination signal.

First, in step S1, it is determined whether or not the flag FMON is set to "1" by a monitor permission determination routine which will be described later. If the answer is negative (NO), the process proceeds to step S9, where the monitor permission determination routine is started again, and the program is terminated. On the other hand, when the answer to step S1 is affirmative (YES), the process proceeds to step S2 to measure the rotation fluctuation of the crankshaft (period fluctuation of the crank angle detection signal), and then executes a misfire determination routine (step S3). Then, it is determined whether or not a flag FMF indicating that the misfire of the engine cylinder has been detected is set to "1" (step S4). That is, it is detected whether or not the engine cylinder has misfired based on the rotation fluctuation of the crankshaft. When the answer is negative (NO), the process proceeds to step S8, while when the answer in step S4 is affirmative (YES), it is determined that one of the cylinders of the engine is misfired. Proceeding to step S5, a misfire cylinder determination routine is executed to determine which cylinder has misfired.

Next, in step S6, it is determined whether or not a predetermined number N of consecutive misfires have occurred. That is, in the event of a continuous misfire, the misfire cylinder is determined each time a TDC determination signal is generated by the misfire cylinder determination routine, as described later, so that the number of consecutive misfires is counted. Here, the predetermined number N
Is set to a value at which it is determined that the engine cylinder has misfired even when the fuel is cut or the engine is stopped, for example, N = 4 or more in the case of a four-cylinder engine. If the answer is affirmative (YES), there is a possibility that the engine 1 is performing fuel cut or the like, and the value obtained by subtracting the predetermined number N from the misfire detection number M is calculated as the misfire number NM.
The predetermined number N is not regarded as misfire as F, and the process proceeds to step S8 to determine the abnormality of the engine. Then, the monitor permission determination routine is restarted (step S1).
0) Terminate this program.

Next, the monitor permission judgment routine and each processing step (subroutine) will be described in sequence.

(1) Monitor Permission Judgment FIG. 4 is a flow chart of the monitor permission judgment routine. This program is executed in the background.

First, the engine water temperature TW (TW sensor 10
Is determined to be equal to or higher than a first predetermined temperature TWL1 (for example, 0 ° C.) (step S11),
When the answer is affirmative (YES), the intake air temperature TA (detected by the TA sensor 9) becomes the predetermined temperature TAL (for example,
0 ° C.) or more. If the answer to either step S11 or step S12 is negative (NO), it is considered that the engine is stopped or immediately after starting, and the routine proceeds to step S24, where the flag FMON is set.
Is set to “0”, and the program ends.

On the other hand, steps S11 and S12
If the answer is affirmative (YES), the process proceeds to step S13, and it is determined whether or not the engine 1 is performing fuel cut. Whether or not the fuel cut is being performed is determined based on the engine speed NE and the valve opening θTH of the throttle valve 3 ′, and is specifically determined by executing a fuel cut determination routine (not shown).

When the answer is affirmative (YES), that is, when the fuel cut is in progress, there is no need to monitor the misfire detection, so the routine proceeds to step S24, where the flag FMON is set to "0" and this program ends. On the other hand, if the answer to step S13 is negative (NO), it is determined whether the lean correction coefficient KLS is equal to or less than "1.0" (step S14). When the answer is affirmative (YES), that is, when KLS <1.0, it is a case where so-called rotational roughness is large and irregular combustion is likely to occur, and the flag FMON is set to “0” in order to prevent erroneous misfire detection. (Step S
24) Terminate this program. On the other hand, step S14
If the answer is negative (NO), it is determined whether or not the vehicle is in a cruise running state (step S15). Here, whether the vehicle is in a cruise running state is, for example, ± 0.8 km
The determination is made based on whether or not the vehicle is in a running state in which the vehicle speed fluctuation within / sec continues for 2 seconds. And the answer is negative (NO)
In the case of, it is determined whether or not the vehicle speed VSP (detected by the vehicle speed sensor 19) is larger than "0". When the answer is affirmative (YES), the vehicle is accelerating or decelerating. The flag FMON is set to "0" (step S24), and this program ends. On the other hand, step S1
When the answer to 6 is negative (NO), it is the case of no load,
The engine speed NE is within a predetermined range (NENL <NE <N
ENH), for example, 750 rpm <NE <3000 rpm
m is determined (step S17). If the answer is negative (NO), the flag FMON is set to "0" (step S24), and the program is terminated.
When the answer is affirmative (YES), the process proceeds to step S18.

On the other hand, the answer in step S15 is affirmative (YE
S), that is, when the vehicle is in the cruise running state, it is determined whether or not the engine coolant temperature TW is higher than a second predetermined temperature TWL2 (step S19). Here, the second predetermined temperature TWL2 is set to a temperature at which irregular combustion due to the rotational roughness of the crankshaft immediately after starting can occur, for example, 20 ° C. If the answer to step S19 is affirmative (YES), it is determined that irregular combustion does not occur, and the process proceeds to step S18. On the other hand, if the answer to step S19 is negative (NO), it is determined whether or not the valve opening θTH of the throttle valve 3 ′ (detected by the θTH sensor 4) is in a fully closed state (step S20). Is positive (Y
ES), the process proceeds to step S24 to set the flag FMO
N is set to "0" and the program ends.

On the other hand, the answer to step S20 is negative (NO).
In other words, when the throttle opening 3TH is not in the fully closed state, it is determined that irregular combustion does not occur, and the process proceeds to step S18. That is, when the engine is not loaded and the cruise travel is being performed and the rotational roughness cannot be generated, the process proceeds to step S18, and it is determined whether the driving state has changed.

Whether or not this operating state has changed is specifically determined based on the flowchart shown in FIG.
This program is also processed in the background in the same manner as the program in FIG.

First, in step S31, the difference between the valve opening degrees of the throttle valve 3'at the time of the previous TDC and the time of the current TDC Δθ.
It is determined whether or not the absolute value of TH is smaller than a predetermined value (for example, 2 °).
Pressure difference ΔPBA of intake pipe absolute pressure PBA at DC
Is determined whether or not the absolute value of is smaller than a predetermined value (for example, 10 mmHg). That is, in steps S31 and S32, it is determined whether or not there is a small change in the driving state that is not detected by the vehicle speed sensor 19. And
If at least one of the answers in step S31 or S32 is negative (NO), it is determined that the operating state has changed, the flag FDRV is set to "1", and the routine returns to the routine in FIG.

On the other hand, steps S31 and S32
If both answers are affirmative (YES), step S33
Then, it is determined whether or not the variation in the intake pipe absolute pressure PBA has been longer than the generation interval of the TDC determination signal. That is, when the fluctuation of the intake pipe absolute pressure PBA detected for each TDC determination signal is small, but the fluctuation amount of the intake pipe absolute pressure PBA increases for a time longer than the generation interval of the TDC determination signal, for example, 100 ms. It is determined whether or not such a variation is greater than a predetermined value. That is, for example, when the transmission 17 is a manual transmission mechanism (MT) and the clutch engagement state is a half-clutch state, or when the transmission 17 is an automatic transmission mechanism (AT) and a shift change is performed. Absolute pressure in intake pipe PBA
Gradually increases. Therefore, in the present embodiment, the fluctuation of the intake pipe absolute pressure PBA is detected at predetermined time intervals longer than the generation interval of the TDC determination signal, and the presence or absence of the fluctuation of the intake pipe absolute pressure PBA is determined. To prevent misfire detection when the vehicle is half-clutched or when an AT vehicle shifts.

Whether or not the intake pipe absolute pressure PBA has changed is specifically determined by executing a PBA variation determination routine shown in FIG. This program is executed by a timer built in the ECU 5 in synchronization with a pseudo signal pulse generated, for example, every 10 ms.

First, in step S51, it is determined whether or not the engine is in the starting mode. Here, whether or not the engine is in the start mode is determined, for example, based on whether or not a starter switch (not shown) of the engine is turned on and whether or not the engine speed is equal to or lower than a predetermined start speed (cranking speed).

Then, the answer in step S51 is affirmative (YE
In the case of S), the timer tmPB is set to a predetermined time T2 (for example, 100 ms) (step S52), and then the variation ΔPBA1 of the intake pipe absolute pressure is set to P
The BA value is set (step S53), the flag FPB is set to "0" (step S54), and the program ends.

On the other hand, when the answer to step S51 is negative (NO) in the subsequent loop, the process proceeds to step S55, and the timer tmPB set in step S52 is set.
It is determined whether or not the timer value of “1” has expired and has become “0”.

When the answer is negative (NO), the program is terminated as it is, while the answer is affirmative (Y
In the case of ES), the step S5 is performed from the PBA value of the current loop.
3 is subtracted from the value ΔPBA1 and the absolute value is set as the variation ΔPBA2 of the absolute pressure in the intake pipe (step S5).
6) Then, the timer value of the timer tmPB is again set to T2 (for example, 10
0 ms) (step S57), and then step S5
3, ΔPBA1 = PBA (step S5)
8) Then, it is determined whether the ΔPBA2 value calculated in step S56 is larger than a predetermined lower limit value ΔPBAL (for example, 40 mmHg) (step S59).

When the answer is negative (NO),
While the flag FPB is set to "0" and the program is terminated, if the answer is affirmative (YES), it is determined that the fluctuation amount of the intake pipe absolute pressure PBA is large, and the flag FPB is determined.
Is set to "1" (step S60), and the routine returns to the routine of FIG.

After performing the PBA fluctuation judgment processing in this way (FIG. 5, step S3), the process proceeds to step S34, and it is judged whether or not the flag FPB is set to "1" by the above-mentioned PBA fluctuation judgment routine. If the answer is affirmative (YES), it is determined that there is a change in the operating state, the flag FDRV is set to "1", and the routine returns to the routine of FIG.

On the other hand, the answer to step S34 is negative (NO).
In step S35, the process proceeds to step S35, in which it is determined whether the load variation ΔAC of the electric device 23 is equal to or greater than a predetermined value ΔACL (for example, 0.8). If the answer is negative (NO), the air conditioner switch It is determined whether there has been an on / off change of 20 (step S36). If the answer is negative (NO), it is determined whether or not the brake light switch 21 has fluctuated on / off (step S3).
7) If the answer is negative (NO), it is determined whether or not the power steering switch has fluctuated on / off (step S38). And the answer is negative (NO)
In the case of, it is further determined whether or not the drive wheels excessively slip and the TCS is operating (step S39).

When the answer is negative (NO),
It is determined whether or not there has been a change in the input / output of the filter (step S40). That is, in the present embodiment, as will be described later, a high-pass filter is used when the engine speed is low, and a low-pass filter is used when the engine speed is high, thereby removing low-frequency and high-frequency noise components. However, in the engine speed NE region near the filter use boundary, for example, in the high-pass filter, the engine speed is 3000 rpm, and in the low-pass filter, the engine speed is 5000 rpm. May fluctuate. Therefore, in the present embodiment, it is determined whether or not such a change in the filter has occurred. If the answer to step S40 is negative (NO), the flag FDRV is set to "0" (step S4).
1) Return to the routine of FIG. That is, step S3
If each of the steps 5 to S40 is negative (NO), it is determined that there is no change in the operating state, the flag FDRV is set to "0", and the routine returns to the routine of FIG.

On the other hand, if at least one of the steps S35 to S40 is affirmative (YES), it is judged that the driving state has changed, and the flag FDRV is set to "1" (step S42). Returning to the routine of FIG.

However, after performing the operation state change determination process as described above (FIG. 4, step S18), step S21.
To determine whether the flag FDRV is set to "1". If the answer is affirmative (YES), the flag FMON is set to "0" (step S).
24) While ending this program, the answer is negative (N
In the case of O), it is determined whether or not the timer tmMON has passed a predetermined time T1 (step S22). When the answer is negative (NO), the process proceeds to step S24, and the flag FMON is set to "1". When the answer is affirmative (YES), the flag FMON is set to "1" and misfire is detected. Is permitted (step S23), and this program ends.

(2) Crankshaft rotation fluctuation measurement (FIG. 3, step S2) FIG. 7 is a flowchart showing a crankshaft rotation fluctuation measurement routine. This program is executed in synchronization with the generation of the TDC discrimination signal. To be done.

First, in step S71, the CRME values from the # 0 stage to the # 6 stage are added based on the mathematical expression (3) to calculate the predetermined rotation time TREV (n).

TREV (n) = CRME (0) + CRME (1) + ... + CRME (6) (3) The ME value for calculating the engine speed NE is from # 0 stage to # 5 as described above. It is calculated by adding the CREM values for 180 ° up to the stage, but the predetermined rotation time T for obtaining the rotation fluctuation amount of the crankshaft
In this embodiment, REV (n) is 210 ° from stage # 0 to stage # 6 in consideration of the effects of misfiring and the like.
It is calculated by adding the CRME value of the minute.

Next, at step S72, it is determined whether or not the engine speed NE is within the use range of the high-pass filter, that is, the engine speed NE is within the range of the lower limit speed NEHPFL and the upper limit speed NEHPFH when the high-pass filter is used. To determine. Here, the lower limit rotational speed NEHPFL
As the rotation speed close to the idle rotation speed, for example, 750r
pm, and the upper limit rotational speed NEHPFH is set to a rotational speed capable of reliably filtering low-frequency components that may be caused by “swinging” of the engine, for example, 3000 rpm.

Then, the answer in step S72 is affirmative (YE
In the case of S), the filter output value FTREV (n) is calculated based on Expression (4) (Step S73).

FTREV (n) = b ₁ TREV (n) + b ₂ TREV (n−1) + b ₃ TREV (n−2) −a ₁ FTREV (n−1) −a ₂ FTREV (n−2) ... ( 4) Here, b ₁ , b ₂ , b ₃ , a ₁ , and a ₂ are filter transfer coefficients, respectively, which are predetermined transfer coefficients for the high-pass filter, for example, b ₁ = 0.2096, b ₂ = 0.4192. , B ₃ =
0.2096, a ₁ = 0.3557, and a ₂ = 0.194.

Also, as an initial value, F in the loop two times before is set.
FTREV (0) which is a TREV value is, as shown in Expression (5), a crankshaft rotation time TREV at the time of the two-last loop.
V (0) is used, and FTREV (1) which is the FTREV value in the previous loop uses the crankshaft rotation time TREV (1) in the previous loop as shown in Expression (6).

FTREV (0) = TREV (0) (5) FTREV (1) = TREV (1) (6) On the other hand, when the answer in step S72 is negative (NO), the process proceeds to step S74, and the engine speed is increased. It is determined whether or not the number NE is within the use range of the low-pass filter, that is, the engine speed NE is within the range between the lower limit rotational speed NELPFL and the upper limit rotational speed NELPFH using the low-pass filter. Here, the lower limit rotational speed NELPFL is set to a rotational speed capable of reliably filtering high frequency noise caused by torsional vibration of the crankshaft or vibration of the journal such as "rattle", for example, 5000 rpm, and an upper limit rotational speed. NELPH as the maximum rotation speed of the engine 1,
For example, it is set to 6500 rpm. When the answer to step S74 is affirmative (YES), the process proceeds to step S75, and the filter output value FTREV (n) is calculated based on the equation (7).

FTREV (n) = b ₁ ′ TREV (n) + b ₂ ′ TREV (n−1) + b ₃ ′ TREV (n−2) −a ₁ ′ FTREV (n−1) −a ₂ ′ FTREV (n -2) (7) Here, b ₁ ′, b ₂ ′, b ₃ ′, a ₁ ′ and a ₂ ′ are filter transfer coefficients, respectively, which are predetermined transfer coefficients for the low pass filter, for example, b ₁ ′ = 0.159, b ₂ ′ = 0.0
_{318, b 3 '= 0.0159,} a 1' = 1.613, a
₂ '= 0.6766.

In addition, as an initial value, F in the two-preceding loop is used.
TREV (0) value and FTREV (1) in previous loop
The values are the TREV (0) value at the time of the previous loop and the TREV (1) at the time of the previous loop, as in the equations (5) and (6).
Set to value.

Next, in step S76, the deviation ΔFTREV (n) between the FTREV (n) value at the current loop and the FTREV (n-1) value at the previous loop is calculated based on the equation (8).

ΔFTREV (n) = FTREV (n) −FTREV (n−1) (8) Next, in step S77, the rotation variation amount ΔΔFTREV (n) when the filter is used is calculated based on Expression (9). Then, this program ends.

[0086]

(Equation 2) That is, when the engine misfires, the deviation ΔFTREV (n) increases, so that the error deviation ΔFTR
It is difficult to accurately detect the rotation fluctuation for detecting misfire only with the EV (n) value. Therefore, in the present embodiment, the rotation fluctuation amount ΔΔFTREV (n) is calculated in consideration of the ΔFTREV value before 3TDC corresponding to one cycle.

On the other hand, the answer to step S74 is negative (NO).
In the case of, the engine speed NE is the mid-range speed (for example,
3000 to 5000 rpm), where no filter is used. Based on equation (10), the crankshaft rotation time TREV (n) in the current loop and the crankshaft rotation time TREV (n-1) in the previous loop Deviation ΔT
REV (n) is calculated.

ΔTREV (n) = TREV (n) −TREV (n−1) (10) Next, in step S79, the same formula (1) is used as the formula (9).
Based on 1), the rotation fluctuation amount ΔΔTREV (n) is calculated in consideration of the ΔFTREV value before 3TDC, and the program ends.

[0089]

(Equation 3) (3) Misfire determination (FIG. 3, step S3) FIG. 8 is a flowchart showing a misfire determination routine. This program is executed in synchronization with generation of a TDC determination signal.

First, in step S81, a misfire determination value map is selected according to the type of the transmission 17 and the clutch engagement state.

That is, as is well known, an automatic transmission (AT)
Has a built-in torque converter for shifting the engine torque and the like, so that the rotation fluctuation is less likely to occur than in the manual transmission mechanism (MT), and the rotation fluctuation amount upon misfiring is smaller than in the manual transmission mechanism. . Therefore, when the same misfire determination value map is used for AT and MT,
For example, when used in AT misfire determination value map based on the MT, misfire determination value is set larger than the desired value, the engine cylinder even though a misfire is a misfire
There is a possibility to be erroneously determined not, whereas when using the determination value map based on the AT to MT misfire determination value is set smaller than a desired value, the engine cylinders have misfired
There is a risk that misfire is erroneously determined to have occurred even though there is no fire. That is, in the case of AT, it is necessary to make the misfire determination value relatively smaller than in the case of MT.

Regarding the engaged state of the clutch, fluctuations in the rotation of the crankshaft are less likely to occur when the clutch is engaged than when the clutch is disengaged.
Therefore, when using the same regardless of the misfire determination value map the engaged state of the clutch, for example, when the clutch is in the engaged state, a misfire determination value is set rather smaller than a desired value, the engine cylinder misfire and there is a possibility that the erroneous determination that it has had no despite misfire, whereas clutch when in the disengaged misfire determination value is set much larger than the desired value, the engine cylinders are misfiring Nevertheless, there is a risk of misjudgment that no misfire has occurred, and it is difficult to perform accurate misfire detection. Therefore, in this embodiment, a different misfire determination value map is selected according to the type of the transmission 17 and the engagement state of the clutch.

Specifically, when the transmission 17 is an automatic transmission (AT), the map selection routine for an AT vehicle shown in FIG.
L) Map or Road / Load (R / L) map is selected.

That is, in step S101, the transmission 1
7 is in the neutral range or the parking range, and if the answer is negative (NO), the R / L map is selected (step S102), while if the answer is affirmative (YES), N / L is selected. After selecting the R map (step S103), the process returns to the routine of FIG.

Further, the transmission 17 is a manual transmission mechanism (MT).
, The MT vehicle map selection routine shown in FIG. 10 is executed to select the MT vehicle N / L map or R / L map.

That is, in step S111, it is determined whether or not the clutch switch is on, and if the answer is affirmative (YES), the N / L map is selected and the process returns to the routine of FIG.

On the other hand, the answer to step S111 is negative (N
O), the vehicle speed VSP is equal to the predetermined value VRL (for example, 5 km
/ h) is determined (step S11).
3). If the answer is negative (NO), it is determined that the vehicle is in the neutral state, and the N / L map is selected (step S112), and the process returns to the routine of FIG.

On the other hand, the answer in step S113 is affirmative (YE
In the case of (S), the R / L map is selected (step S11).
4) Return to the routine of FIG.

After the misfire determination value map is thus selected (FIG. 8, step S81), the process proceeds to step S82 to search the misfire determination value map to calculate the misfire determination value MFDEL.

Specifically, the misfire determination value map is used as the engine speeds NE1 to NE7 and the intake pipe absolute pressures PBA1 and PBA1.
Map values MFDEL1 to MFDEL7 are given according to BA2, and the misfire determination value MFDEL is read out by searching this misfire determination value map or calculated by an interpolation method. That is, an N / L map for an AT vehicle, an R / L map for an AT vehicle, M
Four types of maps, an N / L map for a T car and an R / L map for an MT car, are stored in advance in the storage means 5c, and these predetermined maps selected according to FIGS. 9 and 10 are searched to determine misfire. The value MFDEL is calculated. When the engine speed NE and the intake pipe absolute pressure PBA are the same,
For the above-mentioned reason, the MFDEL value of the map for MT vehicles is AT
It is set to a value larger than the MFDEL value of the car map,
The MFDEL value of the R / L map is the MFDE of the N / L map
The map is created so as to be set to a value larger than the L value.

Next, in step S83, it is determined whether or not a high-pass filter is used when measuring the crankshaft rotation fluctuation. When the answer is affirmative (YES), the amount of rotation fluctuation ΔΔFTREV (n) when using the high-pass filter calculated in step S77 (FIG. 7) is set to “−”.
A value obtained by multiplying “1” and reversing the sign is set as a new rotation fluctuation amount ΔΔFTREV (n). This makes it possible to eliminate the phase difference between when the high-pass filter is used and when the low-pass filter is used, which is convenient for determining a misfiring cylinder described later.

That is, a phase shift of 180 ° occurs between the high-pass filter and the low-pass filter when determining the misfiring cylinder. Therefore, the "shift" of the phase is eliminated by reversing the sign of the rotation fluctuation amount ΔΔTREV (n) when the high-pass filter is used.

Next, the misfire determination value MFDEL is multiplied by the first correction coefficient C1 (for example, 0.5) to obtain the correction determination value MF.
PF is calculated (step S85), and the process proceeds to step S88.

On the other hand, the answer to step S83 is negative (NO).
In the case of, it is determined whether or not a low-pass filter has been used at the time of measuring the rotation fluctuation of the crankshaft (step S86), and if the answer is affirmative (YES), the second correction coefficient C2 (for example, 0.6) is determined. The misfire determination value MFDEL is multiplied to calculate a correction determination value MFPF (step S87), and step S87 is performed.
Go to 88. That is, since the rotation fluctuation amount ΔΔFTREV (n) calculated when the filter is used is smaller than the rotation fluctuation amount ΔΔTREV (n) calculated when the filter is not used, the misfire determination value MFDE set only from the map search is set.
L is small, and there is a risk of misfire detection. Therefore, the first and second correction coefficients C1 and C2 are set to the misfire determination value MFDE
L is multiplied to calculate a correction determination value MFPF, so that misfire can be reliably detected.

Next, in step S88, it is determined whether or not the rotation variation amount ΔΔFTREV (n) when the filter is used is larger than the correction determination value MFPF, and the answer is negative (N
In the case of O), it is determined that no misfire has occurred in the engine cylinder, and the flag FMF is set to "0" (step S89), and this program ends.

On the other hand, the answer in step S88 is affirmative (YE
In the case of S), the misfire of the cylinder is detected, the flag FMF is set to "1" (step S90), and the program ends.

The answer to step S86 is negative (NO).
In other words, when the filter is not used at the time of measuring the rotation fluctuation of the crankshaft, the process proceeds to step S91,
The rotation fluctuation amount ΔΔTREV (n) is equal to the misfire determination value MFDEL.
Determine if it is greater than. When the answer is negative (NO), the flag FMF is set to "0" (step S92). When the answer is affirmative (YES), the misfire of the engine cylinder is detected and the flag FMF is set to "1". "
(Step S93) and terminates the program.

(4) Misfiring Cylinder Discrimination (FIG. 3, Step S5) FIG. 12 is a flow chart of the misfiring cylinder discrimination routine. This program is executed in synchronization with the generation of the TDC discrimination signal.

First, in step S121, it is determined whether or not a filter has been used during the crankshaft rotation fluctuation measurement (see FIG. 7).

When the answer is affirmative (YES), the discriminated cylinder has a phase delay of 180 ° due to the use of the filter, so the cylinder number immediately before the generation of the CYL signal pulse and before the generation of the TDC discrimination signal is detected as a misfiring cylinder. (Step S1
22) End this program.

That is, if a misfire is detected when the first TDC determination signal is generated after the CYL signal pulse is generated, it is determined that # 1CYL is misfired, and the second TD is detected.
If a misfire is detected when the C discrimination signal is generated, # 3CYL
Is determined to have misfired, and when misfire is detected when the third TDC determination signal is generated, # 4CYL is determined to be misfired, and when misfire is detected when the fourth TDC determination signal is generated, # Determine that 2CYL has misfired.

On the other hand, the answer to step S121 is negative (N
In the case of O), the cylinder number corresponding to the generation of the TDC determination signal immediately after the generation of the CYL signal pulse is detected as the misfiring cylinder (step S123), and this program ends.

That is, if a misfire is detected when the first TDC determination signal is generated after the CYL signal pulse is generated, it is determined that # 3CYL is misfired and the second TD is detected.
If a misfire is detected when the C discrimination signal is generated, # 4CYL
Is determined to have misfired, and if misfire is detected when the third TDC determination signal is generated, # 2CYL is determined to be misfired. If misfire is detected when the fourth TDC determination signal is generated, # 2CYL is determined. It is determined that 1 CYL has misfired.

(5) Abnormality Judgment (FIG. 3, Step S8) FIG. 13 is a flowchart of the abnormality judgment routine. This program is executed in synchronization with the TDC judgment signal.

First, in step S131, after the monitor is restarted, a predetermined number of revolutions (for example, 1000 rp) is continuously generated.
It is determined whether or not the crankshaft rotation fluctuation for m) has been measured. That is, after the start of the monitoring, it is determined whether or not the rotation fluctuation of the crankshaft has been measured within the predetermined rotation speed without the fluctuation of the operation state such as the acceleration operation and the fluctuation of the movement of the filter. If the answer is negative (NO), that is, if the operating state fluctuates within a predetermined number of revolutions, the misfire detection process is terminated, while if the answer is affirmative (YES), the process proceeds to step S132, and the misfire is performed. It is determined whether or not the ratio φ is larger than a predetermined value φ ₀ (for example, 0.01). here,
The misfire rate φ is calculated by counting the number of misfires at each predetermined rotation speed, that is, the number of misfire TDCs. For example, when the predetermined rotation speed is 1000 rpm, the TDC determination signal generated by the four-cylinder engine is 2000, and φ ₀ =
In the case of 0.01, it is determined whether or not the misfire rate φ is larger than a predetermined value φ _{0 based on} whether or not the misfire TDC is 20 times or more.

When the answer is negative (NO), the engine is judged to be normal and the program is terminated (step S133). On the other hand, when the answer is affirmative (YES), the engine is judged to be abnormal. , This program ends.

Furthermore, the present invention is not limited to the above embodiment, and, for example, in the above embodiment, the engine cylinder misfires based on the correction judgment value calculated by correcting the misfire judgment value when the filter is used. Although it is determined whether or not (see FIG. 8, step S85, step S86), the rotation fluctuation amount when using the filter may be corrected. That is, FIG.
As shown in the flowchart of FIG.
In step S187, a correction coefficient is applied to the rotation fluctuation amount when the filter is used, that is, a third correction coefficient C3 (for example, 2.0) when the high-pass filter is used, and a fourth correction coefficient C4 (for example, 1.6) when the low-pass filter is used. ) To calculate the corrected rotation fluctuation amount ΔΔFTREVM (n). In step S188, the corrected rotation fluctuation amount ΔFTREVM (n) is calculated.
The same effect can be obtained by determining whether or not is larger than the misfire determination value MFDEL.

In the above embodiment, the filter means is composed of the high pass filter and the low pass filter.
The filter means may be configured by a band-pass filter, and may be configured to calculate the rotation fluctuation amount of the crankshaft by switching to a band-pass filter adapted to each according to the detection result of the engine speed NE.

[0119]

As described in detail above, according to the present invention, the crank angle detecting means for detecting the rotation angle of the crankshaft at every predetermined angle, and the cycle for detecting the periodic fluctuation of the detection signal obtained by the crank angle detecting means. A misfire detection device for an internal combustion engine, comprising: a fluctuation detection means; and a misfire determination means for comparing the detection result of the periodic fluctuation detection means with a predetermined determination value to determine whether or not the internal combustion engine has misfired. Since it has a speed switching mechanism of either one of the mechanism or the manual speed change mechanism, and a judgment value switching means for switching the predetermined judgment value according to the type of the speed switching mechanism,
The predetermined judgment value can be set to an optimum value according to the type of speed switching mechanism, misfire misjudgment can be prevented, and misfire detection accuracy can be improved.

Further, the vehicle has a speed switching mechanism capable of speed switching in a plurality of stages, and the judgment value switching for switching the predetermined judgment value according to the engagement state of a clutch provided between the speed switching mechanism and the engine. Since the means is provided, similarly to the above, the predetermined determination value can be set to an optimum value according to the engagement state of the clutch, misfire misjudgment can be prevented, and misfire detection accuracy can be improved. be able to.

Furthermore, the present invention has a rotation speed detecting means for detecting the engine rotation speed, and a plurality of filter means for outputting only the detection signal in the specific frequency band among the detection signals obtained by the rotation speed detecting means. In addition, since the predetermined judgment value selected by the judgment value switching means is provided with the correction means for reducing the amount of the predetermined judgment value according to the output frequency band of the filter means, a different map is used when the filter is used and when the filter is not used. It is possible to set a predetermined determination value without preparing, and to reduce the load on the control system.

Further, the present invention has a deciding means for deciding whether or not the filter means should be used according to the engine speed detected by the speed detecting means, and
When a misfire is detected by the misfire determining means while the filter means is being used based on the determination result of the determining means, the misfires of the cylinders having the same time delay with respect to the misfire detection timing are sequentially determined. Since the misfiring cylinder discrimination means is provided, the discrimination of the misfiring cylinder is simplified and the discrimination of the misfiring cylinder can be easily performed.

Specifically, when the filter means includes a high-pass filter and a low-pass filter,
Even when the high-pass filter is used, the misfire cylinder can be accurately identified with the same software as when the low-pass filter is used, and the accuracy of misfire detection can be improved and the load on the control system can be reduced.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing one embodiment of a misfire detection device for an internal combustion engine according to the present invention.

FIG. 2 shows a CYL signal pulse, a TDC determination signal, and a CRK signal.
It is a flowchart which shows the generation timing of a signal pulse.

FIG. 3 is a flowchart showing a main routine of the present invention.

FIG. 4 is a flowchart of a monitor permission determination routine.

FIG. 5 is a flowchart of an operating state determination routine.

FIG. 6 is a flowchart of an intake pipe absolute pressure fluctuation determination routine.

FIG. 7 is a flowchart of a crankshaft rotation fluctuation measurement routine.

FIG. 8 is a flowchart of a misfire determination routine.

FIG. 9 is a flowchart of an AT vehicle misfire determination value map selection routine.

FIG. 10 is a flowchart of an MT vehicle misfire determination value map selection routine.

FIG. 11 is an MFDEL map diagram.

FIG. 12 is a flowchart of a misfire cylinder determination routine.

FIG. 13 is a flowchart of an abnormality determination routine.

FIG. 14 is a flowchart showing another embodiment of the misfire determination routine.

[Explanation of symbols]

1 Internal Combustion Engine 5 ECU (cycle fluctuation detecting means, misfire determining means, determination value switching means, rotation speed detecting means, filter means, correcting means, misfire cylinder determining means) 12 CRK sensor (crank angle detecting means)

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location F02P 17/12 F02P 17/00 U

Claims (5)

(57) [Claims] 1. A crank angle detecting means for detecting a rotation angle of a crankshaft for each predetermined rotation angle, a cycle fluctuation detecting means for detecting a cycle fluctuation of a detection signal obtained by the crank angle detecting means, and a cycle fluctuation detecting means. In a misfire detection device for an internal combustion engine, which comprises a misfire determination means for comparing the detection result of the means with a predetermined determination value to determine whether or not the internal combustion engine has misfired, one of an automatic transmission mechanism and a manual transmission mechanism. A misfire detection device for an internal combustion engine, comprising one of the speed switching mechanisms and a determination value switching means for switching the predetermined determination value according to the type of the speed switching mechanism. 2. A rotation speed detecting means for detecting an engine rotation speed, and a plurality of filter means for outputting only a detection signal in a specific frequency band among the detection signals obtained by the rotation speed detecting means, the determination value changeover misfire detecting device according to claim 1 Symbol mounting an internal combustion engine, characterized in that it comprises correction means for decreasing correction according to the output frequency band of the filter means a predetermined judgment value selected by switch means . 3. A crankshaft rotation angle is set at a predetermined rotation angle.
Crank angle detecting means for detecting the
Rotation to detect the engine speed based on the stage detection result
Obtained by the number detecting means and the crank angle detecting means
A cycle variation detecting means for detecting a cycle variation of the detection signal;
By comparing the detection result of the periodic fluctuation detection means with a predetermined judgment value
Misfire determination means for determining whether the internal combustion engine has misfired
In the misfire detecting device for an internal combustion engine with bets and has a speed switching mechanism capable of speed switching in a plurality of stages co
And a clutch provided between the speed switching mechanism and the engine.
Judgment for switching the predetermined judgment value according to the engagement state of the switch
A detection signal obtained by the rotation speed detection means having a value switching means;
A plurality of filters that output only detection signals in a specific frequency band
Filter means and the predetermined value selected by the predetermined value switching means.
The judgment value is reduced according to the output frequency band of the filter means.
And a correction means for correcting the amount.
Fire engine misfire detection device. 4. A determination means for determining whether or not the filter means should be used according to the engine speed detected by the rotation speed detection means, and the filter means is based on the determination result of the determination means. When a misfire is detected by the misfire determination means during use,
The misfire detection device for an internal combustion engine according to claim 2 or 3, further comprising: misfire cylinder determination means for sequentially determining misfires of cylinders having the same time delay with respect to the misfire detection timing. 5. The filter means includes at least a high-pass filter through which frequency components above a first predetermined frequency band pass and a low-pass filter through which frequency components below a second predetermined frequency band pass. The misfire detection device for an internal combustion engine according to claim 3 or 4.