JP4372737B2 - Misfire detection device for internal combustion engine - Google Patents

Description

  The present invention relates to a misfire detection device for an internal combustion engine, and more particularly to a device for determining the presence or absence of misfire based on a rotation speed parameter corresponding to the engine rotation speed.

  Patent Document 1 discloses a technique for determining the presence or absence of misfire based on a feature signal q (n) formed based on a segment time that is a time required for rotation at a predetermined crank angle. The feature signal q (n) is a signal obtained by converting the signal indicating the segment time so as to be indicated as a point on the complex plane, and whether or not misfire has occurred is determined based on the magnitude and phase of the feature signal q (n). Determined.

JP-A-9-119338

  In the conventional method described above, the processing accuracy of the teeth of the crank pulser for detecting the segment time (referred to as “angle transmission vehicle” in Patent Document 1), the variation in characteristics of the magnetic sensor, or the axis of the crankshaft Since the influence of core deflection or the like is not taken into account, there is a possibility that erroneous detection occurs.

  The present invention has been made paying attention to this point, and can suppress the influence of the processing accuracy of the crank pulsar, the variation in the characteristics of the magnetic sensor, the deflection of the axis of the crankshaft, and the like, and can perform accurate misfire detection. An object of the present invention is to provide a misfire detection device for an internal combustion engine.

In order to achieve the above object, the invention according to claim 1 is directed to a misfire detection apparatus for an internal combustion engine that detects misfire of the engine based on a rotation speed parameter (Tx) corresponding to the rotation speed of the internal combustion engine. Frequency component vector calculating means for calculating the frequency component of the parameter (Tx) as a frequency component vector (VX) on the complex plane, and the frequency component vector calculating means during the fuel supply cutoff operation for cutting off the fuel supply to the engine Origin correction vector calculation means for calculating an origin correction vector (Vp) in accordance with the frequency component vector (VX) calculated by the above, and the frequency component vector (VX) calculated during normal operation of the engine as the origin. Correction means for correcting with a correction vector (Vp), and frequency component vector (VX) corrected by the correction means ) And determination means for performing a misfire determination based on the rotational speed of the engine (NE) and the intake pressure (PBA) operating correction vector according to (Vne, and a driving correction vector calculation means for calculating a VTQ), wherein The operation correction vector calculation means calculates a real part component (Rne, Rtq) and an imaginary part component (Ine, Itq) of the operation correction vector according to the rotational speed (NE) and the intake pressure (PBA), respectively. correcting means, said origin correction vector (Vp) and the operating correction vector (Vne, VTQ) by, characterized that you correct the frequency component vector calculated (VX) in normal operation of the engine.

  According to a second aspect of the present invention, in the misfire detection device for an internal combustion engine according to the first aspect, the determination means is configured such that the absolute value (GVXc) of the corrected frequency component vector is greater than a determination threshold (GJUD). It is determined that misfire has occurred, and the misfire cylinder is determined according to the phase angle (φVXc) of the corrected frequency component vector.

According to a third aspect of the present invention, in the misfire detection apparatus for an internal combustion engine according to the first or second aspect , the frequency component vector calculating means receives the rotational speed parameter (Tx) and determines the rotational speed of the engine. A bandpass filter (31) for passing a frequency component in the vicinity of the corresponding center frequency, and calculating a frequency component vector of the rotational speed parameter (Txf) output from the bandpass filter (31). To do.

According to a fourth aspect of the present invention, in the misfire detection device for an internal combustion engine according to any one of the first to third aspects, the frequency component vector calculating means is a frequency component corresponding to the engine speed (NE). The frequency component vector of (X (2)) is calculated.

According to the first aspect of the present invention, the frequency component of the rotation speed parameter is calculated as a frequency component vector on the complex plane, and the frequency component vector calculated during the fuel supply cutoff operation is calculated as the origin correction vector. Is done. An operation correction vector is calculated according to the engine speed and the intake pressure. The frequency component vector calculated during normal operation of the engine is corrected by the origin correction vector and the operation correction vector , and misfire determination is performed based on the corrected frequency component vector. Effects of crank pulser tooth accuracy, angle sensor characteristic variations, or crankshaft axis deflection are reflected in the origin correction vector calculated when the engine is not combusted. By performing the above, it is possible to suppress the influence of the processing accuracy of the crank pulsar, the variation in the characteristics of the magnetic sensor, or the deflection of the axis of the crankshaft, and to perform accurate misfire detection. Further, since the frequency component vector also changes due to changes in engine speed and / or intake pressure, accurate misfire determination can be performed regardless of changes in the engine operating state by applying the operation correction vector together with the origin correction vector. it can.

  According to the second aspect of the present invention, it is determined that misfire has occurred when the absolute value of the corrected frequency component vector is greater than the determination threshold, and the misfire cylinder is determined in accordance with the phase angle of the corrected frequency component vector. A determination is made. Among the frequency components of the rotational speed parameter, by using a specific frequency component that increases when misfire occurs, for example, a frequency component vector of a frequency corresponding to the engine speed, the occurrence of misfire is regarded as an increase in the absolute value of the frequency component vector. Can be detected. Further, since the phase of the frequency component vector changes depending on which cylinder has misfired, the misfiring cylinder can be determined based on the phase of the frequency component vector.

According to the third aspect of the present invention, the rotational speed parameter is band-limited by the band-pass filter that passes a signal having a frequency near the center frequency corresponding to the rotational speed of the engine. It is possible to prevent the frequency component vector from being shifted due to the influence, and perform an accurate misfire determination.

According to the invention described in claim 4 , the frequency component vector of the frequency corresponding to the engine speed is calculated, and the misfire determination is performed based on the frequency component vector. The absolute value of the frequency component vector of the frequency corresponding to the engine rotational speed increases when misfiring continuously occurs in a specific cylinder and a cylinder having an operation phase shifted 360 degrees from the operation phase of the specific cylinder. By paying attention to the vector, it is possible to detect continuous misfires in a plurality of cylinders whose operating phases are shifted by 360 degrees.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram showing a configuration of an internal combustion engine and a control device thereof according to an embodiment of the present invention. An internal combustion engine (hereinafter simply referred to as “engine”) 1 has, for example, four cylinders and includes an intake pipe 2 and an exhaust pipe 5. A throttle valve 3 is provided in the intake pipe 2. The exhaust pipe 5 is provided with a catalytic converter 6 for purifying exhaust gas.

  The fuel injection valve 4 is provided for each cylinder between the engine 1 and the throttle valve 3 and slightly upstream of the intake valve (not shown) of the intake pipe 2, and each injection valve is connected to a fuel pump (not shown). At the same time, it is electrically connected to an electronic control unit (hereinafter referred to as “ECU”) 20 and the opening time of the fuel injection valve 4 is controlled by a control signal from the ECU 20.

An intake pipe absolute pressure (PBA) sensor 11 for detecting the pressure in the intake pipe 2 is provided immediately downstream of the throttle valve 3, and the detection signal is supplied to the ECU 20.
A crank angle position sensor 12 that detects a rotation angle of a crankshaft (not shown) of the engine 1 is connected to the ECU 20, and a signal corresponding to the rotation angle of the crankshaft is supplied to the ECU 20. The crank angle position sensor 12 is a cylinder discrimination sensor that outputs a pulse (hereinafter referred to as “CYL pulse”) at a predetermined crank angle position of a specific cylinder of the engine 1, and relates to a top dead center (TDC) at the start of the intake stroke of each cylinder. A TDC sensor that outputs a TDC pulse at a crank angle position before a predetermined crank angle (every crank angle of 180 degrees in a four-cylinder engine) and one pulse (hereinafter referred to as “CRK”) with a constant crank angle cycle shorter than the TDC pulse (for example, a cycle of 30 °). The CYL pulse, the TDC pulse, and the CRK pulse are supplied to the ECU 20. These pulses are used for various timing controls such as fuel injection timing and ignition timing, and detection of engine speed (engine speed) NE. Further, the ECU 20 detects misfire in the engine 1 based on the CRK pulse generation time interval T30D.

  The ECU 20 shapes input signal waveforms from various sensors, corrects the voltage level to a predetermined level, converts an analog signal value into a digital signal value, a central processing unit (hereinafter referred to as “CPU”). A storage circuit for storing various calculation programs executed by the CPU and calculation results, an output circuit for supplying a control signal to the fuel injection valve 4 and the like. The CPU of the ECU 20 performs misfire detection described below.

Next, the misfire detection method in this embodiment will be described in detail.
FIG. 2 is a time chart showing actual measurement data of the generation time interval T30D. FIG. 2A corresponds to a normal combustion state in which no misfire has occurred, and FIG. 2B shows cylinders # 1 and # 4. Corresponds to a state in which misfires are continuously generated in FIG. 2, and FIG. 6C corresponds to a state in which misfires are continuously generated in the # 2 cylinder and the # 3 cylinder. In this embodiment, since ignition is performed in the order of # 1 cylinder → # 3 cylinder → # 4 cylinder → # 2 cylinder, the ignition timing of the # 1 cylinder and the ignition timing of the # 4 cylinder are 360 in terms of crank angle. The relationship between the ignition timing of the # 2 cylinder and the ignition timing of the # 3 cylinder is the same.

In the present embodiment, a time parameter Tx defined by the following equation (1) is used as a rotation speed parameter corresponding to the rotation speed of the engine 1. In equation (1), k is the discretization time discretized by the time required for rotating the crank angle 30 degrees, and n is the discretization time discretized by the time required for rotating the crank angle 90 degrees.
Tx (n) = T30D (k) + T30D (k-1) + T30D (k-2) (1)

The frequency components X (0) to X (7) of the time parameter Tx are calculated by the following equation (2) using eight detection data Tx (i) (i = n to n-7) of the time parameter Tx. The Expression (2) is an expression for performing discrete Fourier transform. X (0) calculated by equation (2) is a DC component, X (1) is a frequency component (primary component) 0.5 times the frequency fNE corresponding to the engine speed NE, and X (2) is a frequency. fNE frequency component (second order component), X (3) is 1.5 times the frequency component (third order component) of frequency fNE, X (3) is twice the frequency component (fourth order component) of frequency fNE, X (4) is a frequency component (fifth order component) 2.5 times the frequency fNE, X (5) is a frequency component three times the frequency fNE (sixth order component), and X (6) is the frequency component fNE3. Five times the frequency component (seventh order component) and X (7) indicate a frequency component (fourth order component) four times the frequency fNE.

ここで２次成分Ｘ(2)を表す複素数の実部Ｒｘ，虚部Ｉｘを用いて２次成分ベクトルＶＸ(2)を表すと、式（４）のようになる。なお、以下の説明ではＶＸ(2)は、「ＶＸ」と略して記載する。
ＶＸ(2)＝（Ｒｘ，Ｉｘ）
＝（Ｔｘ(0)−Ｔｘ(2)＋Ｔｘ(4)−Ｔｘ(6)，
−Ｔｘ(1)＋Ｔｘ(3)−Ｔｘ(5)＋Ｔｘ(7)） （４） The secondary component X (2) of the formula (2) is given by the following formula (3).

Here, when the secondary component vector VX (2) is expressed using the real part Rx and the imaginary part Ix of the complex number representing the secondary component X (2), the equation (4) is obtained. In the following description, VX (2) is abbreviated as “VX”.
VX (2) = (Rx, Ix)
= (Tx (0) -Tx (2) + Tx (4) -Tx (6),
-Tx (1) + Tx (3) -Tx (5) + Tx (7)) (4)

  FIG. 3 is a diagram showing measurement data of the secondary component vector VX on a complex plane. When no misfire has occurred, the absolute value of the secondary component vector VX is “0” as shown by a white circle. Therefore, the coordinate points corresponding to the secondary component vector VX are distributed near the origin. On the other hand, when misfire occurs continuously in the # 1 cylinder and the # 4 cylinder, the absolute value of the secondary component vector VX increases, and the corresponding coordinate point is indicated by a black circle in the second quadrant. , Distributed in an area away from the origin. When misfires are continuously generated in the # 2 cylinder and the # 3 cylinder, the coordinate points corresponding to the secondary component vector VX are distributed in a region away from the origin in the fourth quadrant.

  Therefore, when the coordinate points corresponding to the secondary component vector VX are distributed outside the discrimination circle CR, it can be determined that a misfire has occurred. Further, it can be determined from the phase angle φVX whether the cylinder where the continuous misfire has occurred is the # 1 cylinder and the # 4 cylinder, or the # 2 cylinder and the # 3 cylinder.

  FIG. 4 is a waveform diagram schematically showing a change in the time parameter Tx when misfire occurs. The # 1 cylinder and the # 4 cylinder continuously misfire in the center of the figure (continuous misfire of the opposed two cylinders). Waveform is shown). Thus, when continuous misfiring of the opposed two cylinders occurs, the frequency component of the frequency fNE corresponding to the engine speed NE, that is, the secondary component increases.

  FIG. 5 is a diagram for explaining a relationship between a misfire pattern and a corresponding frequency component. During normal combustion, combustion occurs twice per revolution of the engine, so the quaternary component increases. Further, when the above-described opposed two-cylinder continuous misfire occurs, the secondary component increases and the primary component also increases slightly. In addition, when one cylinder continuous misfire (for example, a pattern in which only # 1 cylinder continuously misfires) or two cylinder continuous misfire (for example, a pattern in which # 1 and # 3 cylinders continuously misfire) occurs, The primary component becomes large. Therefore, when detecting a one-cylinder continuous misfire or a two-cylinder continuous misfire, the misfire determination is performed based on the frequency component vector corresponding to the primary component.

  The time parameter Tx is calculated from the CRK pulse generation time interval T30D. The CRK sensor that generates the CRK pulse includes a crank pulser (gear) that is fixed to the crankshaft and rotates, and a magnetic sensor that detects the movement of the crank pulser. Consists of. Therefore, the time parameter Tx varies depending on the processing accuracy of the crank pulser, the variation in the characteristics of the magnetic sensor, the axial deflection of the crankshaft (hereinafter referred to as “sensor error factor”), and the like.

  FIG. 6A shows a state in which regions R0, R1, and R2 in which coordinate points corresponding to frequency component vectors are distributed are shifted due to a sensor error factor. In this example, the region R0 corresponds to the normal combustion state, and the regions R1 and R2 correspond to the state where the opposed two-cylinder continuous misfire has occurred. In this example, since a part of the region R1 is inside the determination circle CR, there is a high possibility of erroneous determination as normal combustion despite misfire.

  Therefore, in the present embodiment, an origin correction vector for correcting a shift due to a sensor error factor is calculated, and thereby the frequency component vector VX is corrected. By this correction, as shown in FIG. 6B, the regions R0 to R2 move to the regions R0 'to R2', respectively, and accurate determination can be performed.

  Next, the case where the engine speed NE changes rapidly will be described with reference to FIG. FIG. 7A is a time chart showing the transition of the time parameter Tx during normal combustion. The waveform plotted with a white circle corresponds to the state in which the engine speed NE has not changed, and the waveform plotted with a rectangle is the engine waveform. Corresponding to the state where the rotational speed NE is slowly decelerating, the waveform plotted with a triangle corresponds to the state where the engine rotational speed NE is decelerating rapidly. The coordinate points indicating the frequency component vectors VX corresponding to these waveforms are as shown in FIG. 7B (rectangles and triangles correspond to when decelerating), and the coordinate points are shifted due to the deceleration of the engine speed NE. Therefore, in the present embodiment, a bandpass filter having the frequency component of interest as the center frequency is provided, and the frequency component vector VX is calculated by performing bandpass filter processing on the time parameter Tx. As a result, as shown in FIG. 7C, the coordinate points corresponding to the waveforms shown in FIG. 7A are all located in the vicinity of the origin, and an accurate determination can be made.

  FIG. 8 is a diagram illustrating the distribution of coordinate points corresponding to the frequency component vector. The region indicated by the broken line indicates the distribution of coordinate points when the bandpass filter processing is not performed, and the region indicated by the solid line is the bandpass filter processing. The distribution of coordinate points when As described above, by performing the band-pass filter process, noise components other than the frequency components necessary for misfire detection are attenuated, so that the variation in coordinate points can be reduced. As a result, the accuracy of misfire determination can be improved.

FIG. 9 is a block diagram illustrating the configuration of a misfire determination module that performs misfire determination by the above-described method and executes a failsafe action according to the determination result. The function of the misfire determination module shown in FIG. 9 is actually realized by arithmetic processing by the CPU of the ECU 20.
The misfire determination module shown in FIG. 9 includes a bandpass filter (hereinafter referred to as “BPF”) 31, a frequency component vector calculation unit 32, an engine speed calculation unit 33, a correction vector calculation unit 34, and a determination threshold calculation unit 35. A misfire determination unit 36, a misfire count processing unit 37, and a fail safe processing unit 38.

The BPF 31 performs the bandpass filter processing described with reference to FIGS. 7 and 8 on the time parameter Tx. Specifically, the post-filtering time parameter Txf is calculated by the following equation (5). In the equation (5), a (q) and b (s) are filter coefficients set so that, for example, the gain frequency characteristic becomes the characteristic shown in FIG.

  The frequency component vector calculation unit 32 calculates the frequency component vector (secondary component vector) VX by applying the post-filtering time parameter Txf as Tx in the equation (4). The engine speed calculator 33 calculates the engine speed NE using the average value of the time parameters Tx input during the period of one cycle.

  The correction vector calculation unit 34 calculates a correction vector Vec (= (Rec, Iec)) according to the procedure shown in FIG. 11 according to the engine speed NE, the intake pipe absolute pressure PBA, and the frequency component vector VX.

  In step S11 of FIG. 11, the Rne table shown in FIG. 12A is searched according to the engine speed NE, and the first real part correction value Rne is calculated. The Rne table is set so that the first real part correction value Rne decreases as the engine speed NE increases in the low speed region, and conversely in the high speed region, the engine speed NE increases. The first real part correction value Rne is set to increase. In step S11, an Ine table shown in FIG. 12B is further retrieved according to the engine speed NE to calculate a first imaginary part correction value Ine. The Ine table is set so that the first imaginary part correction value Ine increases as the engine speed NE increases. A vector having correction values Rne and Ine corresponding to the engine speed NE calculated in step S11 as elements is referred to as a rotation speed correction vector Vne (= (Rne, Ine)).

  In step S12, the Rtq table shown in FIG. 12C is retrieved according to the intake pipe absolute pressure PBA to calculate the second real part correction value Rtq. The Rtq table is set so that the second actual part correction value Rtq decreases as the intake pipe absolute pressure PBA increases. In step S12, the Itq table shown in FIG. 12D is further retrieved according to the intake pipe absolute pressure PBA, and the second imaginary part correction value Itq is calculated. The Itq table is set so that the second imaginary part correction value Itq increases as the intake pipe absolute pressure PBA increases. A vector whose elements are correction values Rtq and Itq corresponding to the intake pipe absolute pressure PBA (engine load) calculated in step S12 is referred to as a load correction vector Vtq (= (Rtq, Itq)).

In step S13, it is determined whether or not the fuel cut flag FFC is “1”. The fuel cut flag FFC is set to “1” when a fuel supply cut-off operation for cutting off the fuel supply to the engine 1 is executed in a process (not shown). When the fuel cut flag FFC is “0”, the process immediately proceeds to step S15. When the fuel cut flag FFC is “1” and the fuel supply cut-off operation is being executed, the frequency component vector calculation unit 32 calculates the fuel cut flag FFC. The applied frequency component vector VX (= (Rx, Ix)), the rotation speed correction vector Vne and the load correction vector Vtq calculated in steps S11 and S12 are applied to the following equations (6) and (7), and the real part The origin correction value Rp and the imaginary part origin correction value Ip, that is, the origin correction vector Vp (= (Rp, Ip)) are updated (step S14).
Rp = A * (Rx-Rne-Rtq) + (1-A) * Rp (6)
Ip = A * (Ix-Ine-Itq) + (1-A) * Ip (7)
Here, A is a smoothing coefficient set to a value between 0 and 1, and Rp and Ip on the right side are previously calculated values.

In step S15, a correction vector Vec (= (Rec, Iec)) is calculated by the following equations (8) and (9).
Rec = Rp + Rne + Rtq (8)
Iec = Ip + Ine + Itq (9)

  Returning to FIG. 9, the determination threshold calculation unit 35 searches the GJUD map shown in FIG. 13 according to the engine speed NE and the intake pipe absolute pressure PBA, and calculates the determination threshold GJUD. The GJUD map is set such that the determination threshold GJUD increases as the engine speed NE decreases and the intake pipe absolute pressure PBA increases. That is, in FIG. 13, the lines L1 to L4 indicate that the intake pipe absolute pressure PBA is a predetermined intake pipe absolute pressure PBA1 (for example, 26.7 PA (200 mmHg)), PBA2 (for example, 53.3 (400 mmHg)), PBA3 (for example, 80 PA (600 mmHg)). )), And PBA4 (for example, 106 PA (800 mmHg)), the predetermined intake pipe absolute pressures PBA1 to PBA4 satisfy the relationship PBA1 <PBA2 <PBA3 <PBA4. The determination threshold GJUD is shown in FIG. This is a parameter corresponding to the radius of the discrimination circle CR.

The misfire determination unit 36 performs misfire determination according to the procedure shown in FIG. 14 based on the frequency component vector VX, the correction vector Vec, and the determination threshold GJUD.
In step S21, the frequency component vector VX and the correction vector Vec are applied to the following equation (10) to calculate the correction frequency component vector VXc.
VXc = (Rxc, Ixc)
= VX-Vec = (Rx-Rec, Ix-Iec) (10)

In step S22, the absolute value GVXc of the correction frequency component vector VXc is calculated by the following equation (11). In step S23, the phase angle φVXc of the correction frequency component vector VXc is calculated by the following equation (12).

  In step S24, it is determined whether or not the absolute value GVXc is larger than the determination threshold value GJUD. When the answer is negative (NO), the coordinate point corresponding to the corrected frequency component vector VXc is located inside the determination circle CR. Is determined to be normal combustion (step S25). On the other hand, when GVXc> GJUD, it is determined whether or not the phase angle φVXc is larger than a predetermined angle φR (for example, 90 degrees) and smaller than (φR + 180 degrees) (step S26). If the answer is affirmative (YES), it is determined that continuous misfire has occurred in the # 1 cylinder and the # 4 cylinder (step S27). If the answer to step S26 is negative (NO), it is determined that continuous misfire has occurred in the # 2 cylinder and the # 3 cylinder (step S28).

  Returning to FIG. 9, the misfire count processing unit 37 counts the number of times the misfire determination unit 36 determines that misfire has occurred, and calculates the misfire rate RMF. When the misfire rate RMF exceeds the predetermined misfire rate RMTH, the determination that misfire has occurred is established. When the misfire occurrence determination is confirmed, the fail safe processing unit 38 executes a necessary fail safe action, for example, stopping the fuel supply to the misfire occurrence cylinder.

  As described above in detail, in the present embodiment, the frequency component vector VX is calculated based on the time parameter Tx, and the origin correction vector Vp is calculated according to the frequency component vector VX calculated during the fuel supply cutoff operation. Is done. Then, the frequency component vector VX calculated during the normal operation in which the fuel is supplied is corrected by the origin correction vector Vp, and misfire determination is performed based on the corrected frequency component vector VXc. Therefore, the influence of sensor error factors such as crank pulser processing accuracy, magnetic sensor characteristic variation, or crankshaft axis deflection is suppressed, and accurate misfire detection can be performed.

More specifically, when the absolute value GVXc of the correction frequency component vector VXc is larger than the determination threshold GJUD, it is determined that misfire has occurred, and the misfire cylinder is determined according to the phase angle φVXc of the correction frequency component vector VXc. .
Further, a rotation speed correction vector Vne (correction value Rne, Ine) is calculated according to the engine speed NE, and a load correction vector Vtq (correction value Rtq, Itq) is calculated according to the intake pipe absolute pressure PBA. A misfire determination is performed based on the correction vector Vne, Vtq and the frequency component vector VXc corrected by the origin correction vector Vp. As a result, an accurate misfire determination can be made regardless of changes in the engine operating state. In addition, if the driving | running state which performs misfire determination is limited to a comparatively narrow range, the correction | amendment by correction vector Vne and Vtq (correction value Rne, Ine, Rtq, and Itq) may not be performed.

  In addition, the band limitation of the time parameter Tx is performed by the BPF 31 that passes the frequency component in the vicinity of the frequency corresponding to the engine speed NE, and the frequency component vector VX is calculated using the time parameter Txf after filtering. In addition, the influence of the acceleration / deceleration of the engine 1 and the variation of the frequency component vector VX due to other noise components can be suppressed, and an accurate misfire determination can be performed. Note that the BPF 31 may not be provided when the misfire determination is performed only when the engine 1 is in a steady operation state.

  In the present embodiment, since the misfire determination is performed using the frequency component corresponding to the engine speed NE, that is, the secondary component, as the frequency component vector VX, the opposed two cylinders (a set of the # 1 cylinder and the # 4 cylinder) Or a combination of # 2 cylinder and # 3 cylinder) can be detected, and the misfire cylinder can be determined.

  In the present embodiment, the ECU 20 constitutes a frequency component vector calculation unit, an origin correction vector calculation unit, a correction unit, a determination unit, and an operation correction vector calculation unit. Specifically, the BPF 31 and frequency component vector calculation unit 32 in FIG. 9 correspond to frequency component vector calculation means, steps S13 and S14 in FIG. 11 correspond to origin correction vector calculation means, and step S21 in FIG. Steps S22 to S28 in FIG. 11 correspond to determination means, and steps S11 and S12 in FIG. 11 correspond to operation correction vector calculation means.

  The present invention is not limited to the embodiment described above, and various modifications can be made. For example, in the above-described embodiment, the example in which the opposed two-cylinder continuous misfire is detected by focusing on the secondary component X (2) among the frequency components X (1) to X (7) has been described. As will be apparent from the above, if attention is paid to the primary component X (1), one-cylinder continuous misfire or two-cylinder continuous misfire can be detected by the same method as described above. In this case, the pass band of the BPF 31 is a band having the frequency of the primary component (component corresponding to fNE / 2) as the center frequency. Moreover, the misfire determination may be performed for both the primary component and the secondary component by switching the frequency component of interest in time division.

  Further, instead of the time parameter Tx, a parameter proportional to the time parameter Tx, a reciprocal of the time parameter Tx, or a parameter proportional to the reciprocal may be used as the rotation speed parameter.

  In the above-described embodiment, an example in which the present invention is applied to a four-cylinder engine has been described. However, the number of cylinders of the engine is not limited to this, and the present invention can also be applied to a six-cylinder engine or an eight-cylinder engine. The relationship between the misfire pattern and the frequency component is as shown in FIG. 15A for a 6-cylinder engine and as shown in FIG. 15B for an 8-cylinder engine. As is clear from this figure, as in the case of the four-cylinder engine, the opposed two-cylinder continuous misfire is focused on the secondary component, and the one-cylinder continuous misfire or the two-cylinder continuous misfire is focused on the primary component. Can detect misfire. In both the 6-cylinder engine and the 8-cylinder engine, the relationship between the frequency of the primary component and the engine speed NE is the same as that of the 4-cylinder engine.

  In the above-described embodiment, an example in which the present invention is applied to a gasoline engine has been described. However, the present invention can also be applied to a diesel engine. In that case, as a parameter indicating the engine load, an accelerator pedal depression amount or a fuel injection amount per unit time is used. Furthermore, the present invention can be applied to misfire detection of a marine vessel propulsion engine such as an outboard motor having a crankshaft as a vertical direction.

It is a figure which shows the structure of the internal combustion engine and its control apparatus concerning one Embodiment of this invention. It is a time chart which shows transition of the generation time interval (T30D) of the pulse which occurs every 30 degrees of crank angles. It is a figure which shows distribution of the coordinate point corresponding to the frequency component vector of the specific frequency component of the signal waveform shown in FIG. It is a time chart which shows transition of the time parameter (Tx) according to engine speed. It is a figure for demonstrating the relationship between a misfire pattern and a frequency component. It is a figure for demonstrating the origin correction | amendment of the coordinate point corresponding to a frequency component vector. It is a figure for demonstrating the necessity of the band pass filter process of a time parameter (Tx). It is a figure for demonstrating the effect of a band pass filter process. It is a block diagram which shows the structure of a misfire determination module. It is a figure which shows the gain characteristic of a band pass filter. It is a flowchart which shows the calculation procedure of a correction vector (Vec). It is a figure which shows the table used for calculation of a correction vector. It is a figure which shows the map used for calculation of a determination threshold value (GJUD). It is a time chart which shows the procedure of misfire determination. It is a figure for demonstrating the relationship between the misfire pattern and frequency component in a 6-cylinder engine and an 8-cylinder engine.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Internal combustion engine 11 Intake pipe absolute pressure sensor 12 Crank angle position sensor 20 Electronic control unit 31 Band pass filter (frequency component vector calculation means)
32 Frequency component vector calculation unit (frequency component vector calculation means)
34 Correction vector calculation unit (origin correction vector calculation means, operation correction vector calculation means)
35 determination threshold value calculation unit 36 misfire determination unit (correction unit, determination unit)

Claims (4)

In a misfire detection device for an internal combustion engine that detects misfire of the engine based on a rotational speed parameter corresponding to the rotational speed of the internal combustion engine,
Frequency component vector calculating means for calculating the frequency component of the rotational speed parameter as a frequency component vector on a complex plane;
Origin correction vector calculation means for calculating an origin correction vector according to the frequency component vector calculated by the frequency component vector calculation means during fuel supply cutoff operation for cutting off fuel supply to the engine;
Correction means for correcting the frequency component vector calculated during normal operation of the engine by the origin correction vector;
Determination means for performing misfire determination based on the frequency component vector corrected by the correction means ;
An operation correction vector calculating means for calculating an operation correction vector according to the engine speed and the intake pressure ,
The driving correction vector calculating means calculates a real part component and an imaginary part component of the driving correction vector according to the rotation speed and the intake pressure, respectively.
Said correction means, said origin by correction vector and operation correction vector, the misfire detection apparatus for an internal combustion engine, characterized that you correct the frequency component vectors calculated during normal operation of the engine.   The determination means determines that a misfire has occurred when an absolute value of the corrected frequency component vector is larger than a determination threshold, and determines a misfire cylinder according to a phase angle of the corrected frequency component vector. The misfire detection device for an internal combustion engine according to claim 1. The frequency component vector calculation means includes a band pass filter that receives the rotation speed parameter and passes a frequency component in the vicinity of a center frequency corresponding to the rotation speed of the engine, and the rotation output from the band pass filter. The misfire detection apparatus for an internal combustion engine according to claim 1 or 2 , wherein a frequency component vector of a speed parameter is calculated. The frequency component vector calculation means, the misfire detecting device for an internal combustion engine according to any one of claims 1 to 3 for calculating a frequency component vector of a frequency corresponding to the engine rotational speed.

Images