KR102053321B1 - An engine misfire diagnosis system and method using down sampling and Discrete Fourier Transform - Google Patents

Abstract

The present invention relates to a system and a method for engine misfire diagnosis using downsampling and discrete Fourier transformation. According to an embodiment of the present invention, the system for engine misfire diagnosis capable of diagnosing whether a misfire occurs in a four-cylinder engine by using an output signal of a crankshaft position sensor comprises: a data collection unit to collect output signal data of the crankshaft position sensor during 1 to N cycles; a downsampling unit to downsample the output signal data of the crankshaft position sensor during 1 to N cycles collected by the data collection unit at regular intervals; a discrete Fourier transformation unit to take a discrete Fourier transform of the output signal data of the crankshaft position sensor downsampled by the downsampling unit; a power spectrum density (PSD) calculation unit to calculate power spectrum density based on the discrete Fourier transform of the output signal data of the crankshaft position sensor taken by the discrete Fourier transformation unit; and a misfire occurrence diagnosis unit to diagnose whether a misfire occurs in the engine based on a harmonic order represented by the maximum of the power spectrum density calculated by the PSD calculation unit.

Description

An engine misfire diagnosis system and method using down sampling and Discrete Fourier Transform}

The present invention relates to an engine misfire diagnosis system and method, and more particularly, to down-sample a crankshaft position sensor output signal data, and then perform a discrete Fourier transform and a specific range of the discrete Fourier transformed crankshaft position sensor output signal data. The present invention relates to an engine misfire diagnosis system and method capable of diagnosing the occurrence of misfire in an engine by calculating a power spectral density using only the crankshaft position sensor output signal data.

In order to prevent air pollution caused by the development of the automobile industry, emission regulations have been tightened for a long time. To cope with this, the development of engine electronic control technology has been promoted.

Furthermore, engine electronic control technology is subject to the On Board Diagnosis II (OBD II), which mandates that computers in vehicles have the ability to identify and alert themselves to information required for fault determination. There is a need for a system and method for detecting the possibility of catalyst damage, and when a misfire occurs in an engine cylinder, an engine misfire diagnosis system and diagnosis that identifies whether the misfire has occurred and the type of misfire, and indicates that the engine is abnormal. A method was required.

Accordingly, a method of diagnosing misfire by using engine roughness has been developed since the 1990s and applied to mass production vehicles. Such a method using engine variability takes into account ECU and semiconductor performance levels of the 90s. As a result of development, the cover of the misfire detection region defined by CARB is limited, but there is a problem that it is not possible to accurately diagnose misfire in some regions such as high RPM and low load intervals.

In addition, in recent years, in addition to the method using engine variability, a method of diagnosing misfire by measuring the ion current generated in the spark plug circuit during explosion process or diagnosing misfire by measuring the combustion pressure directly has been proposed. The above methods have caused a drastic increase in automobile prices due to the addition of new functions or the addition of new sensors to existing vehicles, which has been limited in mass production.

In order to solve this problem, the frequency analysis method of detecting an engine misfire by discrete Fourier transforming the output signal measured from the crankshaft position sensor without additional sensor or equipment has been steadily reviewed. Since it is a method of diagnosing misfire based on the output signal of the crankshaft position sensor, in order to diagnose misfire, the signal data corresponding to the number of teeth formed on the outer circumferential surface of the flywheel had to be analyzed. There was a problem that the computational load on the ECU was increased due to the large amount of computation to be calculated.

That is, even in the case of the conventional frequency analysis method, there is a limitation that the misfire diagnosis speed and precision are reduced due to the computational load of the ECU caused by the increased computation amount, and thus the computation load can be lowered while maintaining the precision of the engine misfire diagnosis. A new engine misfire diagnosis method is required.

US Patent Publication No. 7530261 “Fourier-based misfire detection strategy”

The technical problem to be achieved by the present invention is not limited to the technical problem mentioned above, and other technical problems not mentioned above may be clearly understood by those skilled in the art from the following description. will be.

In order to solve the above problems, the present invention is a system capable of diagnosing the occurrence of misfire in a four-cylinder engine using the output signal of the crankshaft position sensor in one embodiment, the crankshaft position sensor for 1 to N cycles A data collector for collecting output signal data of the data; A down sampling unit for down sampling the crankshaft position sensor output signal data for 1 to N cycles collected by the data collection unit at regular intervals; A discrete Fourier transform unit for performing discrete Fourier transform on the down-sampled crankshaft position sensor output signal data; A PSD calculator configured to calculate a power spectral density (PSD) based on the crankshaft position sensor output signal data discretely Fourier transformed by the discrete Fourier transform unit; And a misfire occurrence diagnosis unit for diagnosing whether misfire has occurred in the engine based on a harmonic order value in which the maximum value of the power spectral density calculated by the PSD calculator is displayed (where N is a natural number). to be.).

In this case, the down sampling unit down-samples the crankshaft position sensor output signal data for 1 to N cycles collected by the data collection unit above the Nyquist sampling criterion.

Particularly, the down sampling unit down-samples the crankshaft position sensor output signal data during 1 to N cycles collected by the data collecting unit to a level of 1/6 or more and 1/2 or less.

In addition, the PSD calculator is characterized in that only calculates the power spectral density value when the harmonic order value is 4N or less.

In addition, the misfire occurrence diagnosis unit may diagnose that misfire in the engine occurs when the maximum value of the power spectral density appears at a value whose harmonic order value is not 4N.

According to another aspect of the present invention, there is provided a method for diagnosing whether misfire in a four-cylinder engine is generated by using an output signal of a crankshaft position sensor, wherein the crankshaft position sensor is used for 1 to N cycles. A data collection step of collecting output signal data of the; A down sampling step of down sampling the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step at regular intervals; A discrete Fourier transform step of performing discrete Fourier transform on the down-sampled crankshaft position sensor output signal data; A PSD calculation step of calculating a power spectral density (PSD) based on the crankshaft position sensor output signal data which is discrete Fourier transformed in the discrete Fourier transforming step; And a misfire occurrence diagnosis step of diagnosing whether misfire in the engine occurs based on a harmonic order value in which the maximum value of the power spectral density calculated in the PSD calculation step is represented, provided that N is a natural number. to be.).

At this time, the down sampling step is characterized in that down-sample the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step above the Nyquist sampling reference.

In particular, the down sampling step is characterized by down-sampling the crankshaft position sensor output signal data during 1 to N cycles collected in the data collection step to 1/6 or more and 1/2 or less.

In the PSD calculation step, only the power spectral density value when the harmonic order value is 4N or less is calculated.

In addition, the misfire occurrence diagnosing step is characterized in that the misfire in the engine is diagnosed when the maximum value of the power spectral density appears at a value other than the harmonic order value of 4N.

The engine misfire diagnosis system and method of the present invention can reduce the amount of computation that must be calculated in the ECU by down-sampling the output signal of the crankshaft position sensor based on the Nyquist sampling theory and then performing discrete Fourier transform. The computational load on the ECU can be reduced.

In addition, the engine misfire diagnosis system and method of the present invention can diagnose the engine misfire only by the power spectral density value when the harmonic order value is 4N or less, thereby reducing the calculation speed required for misfire diagnosis.

1 is a conceptual diagram of an engine misfire diagnosis system according to an embodiment of the present invention.
2 is a flowchart of an engine misfire diagnosis system according to an embodiment of the present invention.
3 is a view for explaining a process of down sampling the crankshaft position sensor output signal data in the down sampling unit of the present invention.
4 is a diagram showing discrete Fourier transform results of crankshaft position sensor output signal data before and after downsampling.
5 is a graph comparing the harmonic order value in which the maximum value of the power spectral density of the crankshaft position sensor output signal data varies depending on whether misfire in the engine occurs.
6 is a flowchart illustrating an engine misfire diagnosis method according to another embodiment of the present invention.

Hereinafter, with reference to the drawings will be described the present invention in more detail. It should be noted that the same elements in the figures are represented by the same numerals wherever possible. In addition, detailed description of well-known functions and configurations that may unnecessarily obscure the subject matter of the present invention will be omitted.

When a component is referred to as being connected or connected to another component, it is to be understood that although the component may be directly connected or connected to the other component, other components may exist in the middle. In addition, when a member is located "on" another member throughout this specification, this includes not only when a member is in contact with the other member, but also when there is another member between the two members.

In the present application, the term “comprises” or “having” is intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more. It is to be understood that it does not exclude in advance the possibility of the presence or addition of other features or numbers, steps, operations, components, components or combinations thereof.

As described above, a frequency analysis method for diagnosing engine misfire has been proposed using a discrete Fourier transformed crankshaft position sensor output signal (Tooth time signal), but the conventional frequency analysis method is a flywheel. Signal data corresponding to the number of teeth formed on the outer circumferential surface of (10) had to be analyzed.

For example, according to the conventional frequency analysis method, when 60 teeth are formed on the outer circumferential surface of the flywheel 10, 60 crankshaft position sensor 20 output signals must be analyzed for engine misfire diagnosis. did.

That is, the conventional frequency analysis method has a problem that the amount of data to be analyzed is greatly increased as the number of data to be analyzed increases according to the number of teeth formed on the outer circumferential surface of the flywheel 10, and thus the calculation applied to the ECU As the load increases, calculation speed may be slowed or a situation in which misfire diagnosis may not be possible may be limited.

Accordingly, the present invention introduces the Nyquist's Sampling theory to the engine misfire diagnosis process to downsample the crankshaft position sensor 20 output signal, and to downsample the crankshaft position sensor output signal 20. By providing an engine misfire diagnosis system and method capable of diagnosing engine misfire based on the above, it is intended to overcome the limitations of the conventional frequency analysis method.

Hereinafter, with reference to FIGS. 1 to 5, the engine misfire diagnosis system according to an embodiment of the present invention will be described.

1 is a conceptual diagram of an engine misfire diagnosis system according to an embodiment of the present invention, FIG. 2 is a flowchart of an engine misfire diagnosis system according to an embodiment of the present invention, and FIG. 3 is a crankshaft in a down sampling unit of the present invention. A diagram for describing a process of down sampling the position sensor output signal data.

4 is a diagram showing discrete Fourier transform results of crankshaft position sensor output signal data before and after downsampling, and FIG. 5 shows a maximum power spectral density of crankshaft position sensor output signal data depending on whether misfire in the engine occurs. It is a graph comparing the harmonic order value that appears.

According to an embodiment of the present invention, in a system capable of diagnosing whether misfire occurs in a four-cylinder engine by using an output signal of the crankshaft position sensor 20, the output of the crankshaft position sensor 20 for 1 to N cycles is determined. A data collector 100 collecting signal data and down sampling down the crankshaft position sensor 20 output signal data for 1 to N cycles collected by the data collector 100 at regular intervals. Through a discrete Fourier transform unit 400 and a discrete Fourier transform unit 400 for discrete Fourier transforming the crankshaft position sensor 20 output signal data down-sampled by the down sampler 200. The PSD calculator 500 calculates a power spectral density PSD based on the discrete Fourier transformed crankshaft position sensor 20 output signal data, and the power spectral density calculated by the PSD calculator 500. Provides an engine misfire diagnostic system that includes a harmonic daegap this order (order Harmonic) misfire occurrence diagnosis unit 600 for diagnosing whether or not a misfire occurs in the engine based on the value indicated.

At this time, the four-cylinder engine is composed of a first cylinder, a second cylinder, a third cylinder, a fourth cylinder, wherein the first cylinder, the second cylinder, the third cylinder, the fourth cylinder is a first cylinder-a third Light is fired in order of cylinder-fourth cylinder-second cylinder. In addition, the N is a natural number, it means the number of cycles that can be diagnosed at the same time, the N is preferably 1, which is limited to the case of 1 being N because it is only a preferred embodiment of the present invention It is not.

The data collector 100 constituting the engine misfire diagnosis system of the present invention collects output time data of the crankshaft position sensor 20, and the crank collected by the data collector 100. The axis position sensor 20 output signal data is transmitted to the down sampling unit 200 again.

Next, the down sampling unit 200 corresponding to the core component of the engine misfire diagnosis system of the present invention is the crankshaft position collected by the data collection unit 100 based on the Nyquist's Sampling theory. The sensor 20 serves to downsample the output signal data.

)을 알 수 있다. 이 때,

은 샘플링 주파수를 의미하고,

는 엔진의 주파수를 의미한다.Here, the Nyquist sampling theory means that all signals are sampled at regular intervals at a frequency corresponding to twice the highest frequency included in the signal, so that the signal can be perfectly recorded before sampling. When applied to engine misfire diagnosis systems, sampling frequency must be at least twice the engine frequency to achieve engine speed without signal aliasing.

Can be seen. At this time,

Means the sampling frequency,

Is the frequency of the engine.

In addition, when the down sampling unit 200 downsamples the crankshaft position sensor 20 output signal data using a low pass filter, the down sampling unit 200 should downsample at a predetermined interval. For example, when downsampling the crankshaft position sensor 20 output signal data to 1/3 level, it should be used as one signal for every third data.

Engine frequency

)Sampling frequency (

) RPM

One 1/2 1/3 1/4 ··· 1/15 1500 50 100 400 1500 750 500 375 ··· 100 3000 100 200 800 3000 1500 1000 750 ··· 200 4000 133 267 1067 4000 2000 1333 1000 ··· 267 5000 167 333 1333 5000 2500 1667 1250 ··· 333 6000 200 400 1600 6000 3000 2000 1500 ··· 400

(단, L은 엔진의 기통수)로 표현될 수 있다.At this time, the engine frequency is determined according to the engine speed and the engine cylinder, the engine frequency according to the engine speed and engine cylinder is

(Where L is the number of cylinders of the engine).

During the four strokes of suction, compression, explosion, and exhaust, the rotational speed of the crankshaft changes periodically, and the crankshaft position sensor 20 can detect such periodic rotational speed based on the flywheel 10. .

)를 산출할 수 있는 바, In addition, in the output signal of the crankshaft position sensor 20, a main signal may appear in a specific frequency band, and the frequency of the engine may be found by finding a frequency band indicated by the main signal.

) Can be calculated,

)을 토대로 상기 상기 데이터 수집부(100)에서 수집된 1부터 N 사이클 동안의 크랭크축 위치 센서 출력 신호 데이터를 나이퀴스트 샘플링 기준 이상으로 다운 샘플링할 수 있다. 여기서, 나이퀴스트 샘플링 기준은 앞서 언급한 샘플링 주파수는 엔진 주파수의 2배 이상이 되어야 한다는 것을 의미한다.The down sampling unit 200 of the present invention is the frequency of the engine calculated through the above process (

), Down-sampled the crankshaft position sensor output signal data during 1 to N cycles collected by the data collector 100 above the Nyquist sampling criterion. Here, the Nyquist sampling criterion means that the aforementioned sampling frequency should be more than twice the engine frequency.

)는 엔진의 기통수와 엔진 회전 속도(RPM)에 따라 달라질 수 있으며, 나이퀴스트 샘플링 이론에 따르면 신호 왜곡 없이 엔진 속도를 구현하기 위해서는 샘플링 주파수(

)가 엔진 주파수의 2배(

) 이상은 되어야 하는 바, 상기 다운 샘플링부(200)는 나이퀴스트 샘플링 기준에 맞추어 크랭크축 위치 센서(20) 출력 신호를 최대 1/15 수준까지 다운 샘플링할 수 있다.Referring to <Table 1> with reference to the down sampling process of the down sampling unit 200 in more detail, the engine frequency (

) Depends on the number of cylinders in the engine and the engine speed (RPM), and according to the Nyquist sampling theory, to achieve engine speed without signal distortion,

) Is twice the engine frequency (

The down sampling unit 200 may downsample the crankshaft position sensor 20 output signal up to 1/15 level in accordance with the Nyquist sampling criteria.

At this time, the down sampling unit 200 selects the crankshaft position sensor 20 output signal data for 1 to N cycles collected by the data collection unit 100 at regular M intervals, as shown in FIG. 3. And down-sampling the crankshaft position sensor 20 output signal up to a level of 1 / M (however, M is a natural number of 15 or less).

In addition, the down sampling unit 200 does not simply select the crankshaft position sensor 20 output signal data at regular M intervals, but calculates an average value of the M crankshaft position sensor 20 output signal data. The crankshaft position sensor 20 output signal may be down sampled up to a level of 1 / M by storing the calculated average value.

)가 엔진 주파수의 8배(

) 수준인 것이 바람직하다고 알려져 있는 바, 상기 다운 샘플링부(200)는 상기 데이터 수집부(100)에서 수집된 1부터 N 사이클 동안의 크랭크축 위치 센서(20) 출력 신호 데이터를 1/3 수준으로 다운 샘플링하는 것이 바람직하나, As described above, the down-sampling unit 200 can downsample the crankshaft position sensor 20 output signal up to a level of 1/15 based on the Nyquist sampling criterion.

) Is eight times the engine frequency (

It is known that the level of the down sampling unit 200, the crankshaft position sensor 20 output signal data for 1 to N cycles collected by the data collection unit 100 to 1/3 level Down sampling is preferred,

)를 엔진 주파수의 8배(

) 수준보다 낮게 선정하더라도 무관하므로, 상기 다운 샘플링부(200)는 상기와 같은 엔진 실화 진단 과정의 특성을 고려하여 상기 데이터 수집부(100)로부터 수집된 1부터 N 사이클 동안의 크랭크축 위치 센서(20) 출력 신호 데이터를 1/6 이상 1/2 이하 수준으로 다운 샘플링하는 것이 가장 바람직하다.Unlike the general downsampling process, in the engine misfire diagnosis process, the output signal of the crankshaft position sensor 20 in the time domain is not converted into the frequency domain and then again converted into the time domain.

) Is eight times the engine frequency (

Since it is irrelevant even if it is selected to be lower than the level, the down sampling unit 200 takes into account the characteristics of the engine misfire diagnosis process as described above, and the crankshaft position sensor (1 to N cycles) collected from the data collection unit 100 for 1 to N cycles. 20) It is most preferable to downsample the output signal data to the level of 1/6 or more and 1/2 or less.

Next, the engine misfire diagnosis system of the present invention includes a signal processing unit 300 and the number of teeth formed on the outer circumferential surface of the flywheel 10 during 1 to N cycles downsampled by the down sampling unit 200. After receiving the output signal (Tooth time signal) of the crankshaft position sensor 20 by 2 times, the output signal of the crankshaft position sensor 20 received in the course of rotating the crankshaft 720 ° per cycle Since it is redundant, half of the output signal data of the received crankshaft position sensor 20 is filtered through the filter.

In addition, the signal processor 300 finally filters the output signal data of the crankshaft position sensor 20 filtered through the above process through a filter to finally filter only data up to a preset harmonic order.

For example, when the preset harmonic order is 6, the signal processing unit 300 receives the output signal of the crankshaft position sensor 20 through a filter process from the harmonic order of 1 to 6 crankshaft. The output signal of the position sensor 20 can be reduced, thereby simplifying the calculation process in the ECU and improving the calculation speed.

Next, the Discrete Fourier Transformer 200 of the present invention is one to N cycles processed by the signal processor 300 (wherein, one cycle is a section in which the crankshaft rotates by 720 °, and N cycles of the crankshaft The output signal of the crankshaft position sensor 20 using the output signal data of the crankshaft position sensor 20 and the angular velocity of the crankshaft (angular velocity of the flywheel). Discrete Fourier Transform (DFT) is used to transform the frequency domain.

As mentioned above, in one embodiment, the crankshaft position sensor 20 output signal data can be downsampled to a third level, and the discrete Fourier transformation is performed without downsampling the crankshaft position sensor 20 output signal data. Since the results of the discrete Fourier transform after down-sampling the crankshaft position sensor 20 output signal data to a third level are the same as those illustrated in FIG. 4, the downsampler ( Discrete Fourier transforming the output signal of the crankshaft position sensor 20 during 1 to N cycles downsampled in 200 does not affect the precision of the engine misfire diagnosis.

In this case, since the process of discrete Fourier transforming the output signal of the crankshaft position sensor 20 is the same as the conventionally known method, a detailed description of the discrete Fourier transforming process of the discrete Fourier transform unit 400 is omitted in the present invention. do.

Next, the PSD calculator 500 calculates a power spectral density based on the output signal data of the crankshaft position sensor 20 discretely Fourier transformed by the Discrete Fourier Transformer 400.

In this case, the power spectral density means the energy density per unit frequency, and the power spectral density value of the crankshaft position sensor 20 output signal data calculated by the PSD calculator 500 to the misfire occurrence diagnosis unit 600. Can be used to diagnose misfire occurrence in the engine misfire diagnosis system of the present invention.

Lastly, as described above, the misfire occurrence diagnosis unit 600 diagnoses misfire in the engine based on the power spectral density calculated by the PSD calculator 500. The engine misfire diagnosis process of the generation diagnosis unit 600 will be described in more detail.

As a result of the repeated test, when the normal combustion in the engine is made, the maximum value of the power spectral density for the crankshaft position sensor 20 output signal for 1 to N cycles appears when the harmonic order is 4N (k = 4N). Otherwise, when misfire occurs in some cylinders in the engine, it was confirmed that the maximum value of the power spectral density for the crankshaft position sensor 20 output signal appears when the harmonic order is not 4N (k ≠ 4N). In diagnosing engine misfire based on the crankshaft position sensor 20 output signal data for 4 cycles, the harmonic order is 16 (k) as shown in FIG. = 4 X 4 = 16), the maximum value of the power spectral density appears, and ii) when misfire occurs in some cylinders in the engine, as shown in FIG. When not 16 (k ≠ 16), the maximum value of the power spectral density appears.

The misfire occurrence diagnosis unit 600 according to the present invention has a harmonic order value in which the maximum value of the power spectral density with respect to the output signal of the crankshaft position sensor 20 for 1 to N cycles is displayed according to whether misfire has occurred as described above and the test result. Note that this is different, i) when the harmonic order value is 4N, if the maximum value of the power spectral density appears, the engine is diagnosed as normal combustion; and ii) the power spectral density when the harmonic order value is not 4N. If the maximum value of appears, it can be diagnosed that a misfire has occurred in the engine.

As described above, the misfire occurrence diagnosis unit 600 of the present invention can diagnose whether misfire has occurred in the engine by using only the power spectral density value when the harmonic order value is 4N or less. Preferably, the PSD calculator 500 calculates only a power spectral density value when the harmonic order value is 4N or less.

Since the power spectral density value when the harmonic order value is larger than 4N does not affect the engine misfire diagnosis, even if only the power spectral density value when the harmonic order value is 4N or less is calculated as described above, the diagnosis of misfire in the engine is performed. It doesn't matter.

Next, the engine misfire diagnosis method according to another embodiment of the present invention will be described with reference to FIG. 6.

6 is a flowchart of an engine misfire diagnosis method according to another embodiment of the present invention, and the detailed description of each process (step) of the engine misfire diagnosis method according to the present invention is the same as described above, The description will be omitted.

According to another embodiment of the present invention, a method for diagnosing whether misfire in a four-cylinder engine is generated by using an output signal of a crankshaft position sensor includes collecting output signal data of the crankshaft position sensor for 1 to N cycles. Down sampling step (S220) for down sampling the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step (S210) and the data collection step (S210) at regular intervals, the down Discrete Fourier Transform step (S230) for converting the down-sampled crankshaft position sensor output signal data in the sampling step (S220), the discrete Fourier transformed crankshaft position sensor output signal data in the discrete Fourier transform step (S230) PSD calculation step (S240) for calculating a power spectral density (PSD) on the basis of the power calculated in the PSD calculation step (S240) On the basis of the harmonic order (Harmonic order) value up to the value that appears in the spectrum density provides an engine misfire diagnostic method (S200) comprising a misfire has occurred diagnostic step (S250) for diagnosing whether or not a misfire occurs in the engine.

At this time, the down sampling step (S220) is to down-sample the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step (S210) above the Nyquist sampling reference, the crankshaft position You can downsample the sensor output signal data up to 1/15,

)가 엔진 주파수의 8배(

) 수준이 되도록 다운 샘플링하는 것이 낫다고 알려져 있는 바, 상기 다운 샘플링 단계(S220)는 상기 데이터 수집 단계(S210)에서 수집된 1부터 N 사이클 동안의 크랭크축 위치 센서 출력 신호 데이터를 1/3 수준으로 다운 샘플링하는 것이 바람직하나,In practical terms, the sampling frequency (

) Is eight times the engine frequency (

It is known that it is better to downsample to a level of), and the downsampling step (S220) causes the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step (S210) to 1/3 level. Down sampling is preferred,

)를 엔진 주파수의 8배(

) 수준보다 낮게 선정하더라도 무관하므로, 다운 샘플링 단계(S220)에서는 상기와 같은 엔진 실화 진단 과정의 특성을 고려하여 상기 데이터 수집 단계(S210)로부터 수집된 1부터 N 사이클 동안의 크랭크축 위치 센서(20) 출력 신호 데이터를 1/6 이상 1/2 이하 수준으로 다운 샘플링하는 것이 가장 바람직하다.Unlike the general downsampling process, the engine misfire diagnosis process does not need to convert the crankshaft position sensor output signal in the time domain into the frequency domain and then convert it back into the time domain.

) Is eight times the engine frequency (

Since it is irrelevant even if it is selected below the level, the down-sampling step (S220) takes into account the characteristics of the engine misfire diagnosis process as described above, and the crankshaft position sensor (20) for 1 to N cycles collected from the data collection step (S210). It is most preferable to downsample the output signal data to a level of 1/6 or more and 1/2 or less.

In addition, when normal combustion occurs in the engine, the maximum value of the power spectral density for the crankshaft position sensor output signal for 1 to N cycles appears when the harmonic order is 4N (k = 4N), and is misfired in some cylinders of the engine. In the case of, the maximum value of the power spectral density for the crankshaft position sensor output signal appears when the harmonic order is not 4N (k ≠ 4N).

In the misfire occurrence diagnosis step (S250), i) when the harmonic order value is 4N, when the maximum value of the power spectral density appears, the engine is diagnosed as normal combustion, and ii) when the harmonic order value is not 4N. If the maximum value of the spectral density appears, it can be diagnosed that misfire in the engine has occurred.

In addition, as described above, it is possible to diagnose whether misfire has occurred in the engine by using only a power spectral density value of 4N or less harmonic order in the misfire occurrence diagnosis step (S250). By calculating only the power spectral density values below, it is possible to reduce the amount of calculation of the ECU and improve the ECU calculation speed.

In summary, the engine misfire diagnosis system and method of the present invention apply Nyquist sampling theory to the engine misfire diagnosis process to reduce the amount of ECU computation, thereby improving the ECU calculation speed and reducing the computational load on the ECU. In addition, by presenting a desirable down-sampling criterion in the engine misfire diagnosis process using the frequency analysis method, it can contribute to improving the practicality of the frequency analysis method.

In the present invention, after downsampling in a four-cylinder engine, the process of diagnosing a misfire has been described. However, the engine misfire diagnosis process of the present invention is not limited to a four-cylinder engine. Naturally, it can be extended to various types of engines.

While the above has been shown and described with respect to preferred embodiments and applications of the present invention, the present invention is not limited to the specific embodiments and applications described above, the invention without departing from the gist of the invention claimed in the claims Various modifications can be made by those skilled in the art, and these modifications should not be individually understood from the technical spirit or the prospect of the present invention.

Also, the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise.

The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

100: data collection unit
200: down sampling unit
300: signal processing unit
400: Discrete Fourier Transform
500: PSD calculation unit
600: misfire occurrence diagnosis unit

Claims (10)

In the system capable of diagnosing the occurrence of misfire in the four-cylinder engine using the output signal of the crankshaft position sensor,
A data collector configured to collect output signal data of the crankshaft position sensor for 1 to N cycles;
A down sampling unit for down sampling the crankshaft position sensor output signal data for 1 to N cycles collected by the data collection unit at regular intervals;
A discrete Fourier transform unit for performing discrete Fourier transform on the down-sampled crankshaft position sensor output signal data;
A PSD calculator configured to calculate a power spectral density (PSD) based on the crankshaft position sensor output signal data discretely Fourier transformed by the discrete Fourier transform unit; And
A misfire occurrence diagnosis unit for diagnosing whether misfire has occurred in the engine based on a harmonic order value in which the maximum value of the power spectral density calculated by the PSD calculator is displayed;
Including,
The misfire occurrence diagnosis unit,
An engine misfire diagnosis system comprising diagnosing a misfire in an engine when the maximum value of the power spectral density appears when the harmonic order value is not 4N. (N is a natural number.)
The method of claim 1,
The down sampling unit,
And down-sample the crankshaft position sensor output signal data during the 1 to N cycles collected by the data collection unit above the Nyquist sampling criterion.
The method of claim 2,
The down sampling unit,
The engine misfire diagnosis system, characterized in that the down-sample the crankshaft position sensor output signal data for 1 to N cycles collected by the data collection unit to 1/6 or more and 1/2 or less.
The method of claim 1,
The PSD calculation unit,
The engine misfire diagnosis system, wherein only the power spectral density value is calculated when the harmonic order value is 4N or less.
delete In the method for diagnosing the occurrence of misfire in the four-cylinder engine using the output signal of the crankshaft position sensor,
A data collection step of collecting output signal data of the crankshaft position sensor for 1 to N cycles;
A down sampling step of down sampling the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step at regular intervals;
A discrete Fourier transform step of performing discrete Fourier transform on the down-sampled crankshaft position sensor output signal data;
A PSD calculation step of calculating a power spectral density (PSD) based on the crankshaft position sensor output signal data which is discrete Fourier transformed in the discrete Fourier transforming step; And
A misfire occurrence diagnosis step of diagnosing the occurrence of misfire in the engine on the basis of the harmonic order value in which the maximum value of the power spectral density calculated in the PSD calculation step appears;
Including,
The misfire occurrence diagnosis step,
An engine misfire diagnosis method comprising diagnosing a misfire in an engine when the maximum value of the power spectral density appears when the harmonic order value is not 4N. (N is a natural number.)
The method of claim 6,
The down sampling step,
And down-sample the crankshaft position sensor output signal data for 1 to N cycles collected in the data collection step above the Nyquist sampling criterion.
The method of claim 7, wherein
The down sampling step,
And down-sample the crankshaft position sensor output signal data during 1 to N cycles collected in the data collection step to 1/6 or more and 1/2 or less.
The method of claim 6,
The PSD calculation step,
An engine misfire diagnosis method comprising calculating only the power spectral density value when the harmonic order value is 4N or less.
delete

Images