KR102473710B1 - Method and apparatus for fuel amount training and compensation in GDI engine - Google Patents

Abstract

An invention related to fuel amount correction and learning thereof in a catalytic heating injection low-volume section where engine misfires due to engine roughness may occur most frequently. The present invention corrects the roughness generated by the flow rate deviation of the injector in the low flow rate section of the catalytic heating injector for each cylinder to prevent misfiring. ) value to calculate a fuel amount correction factor, which is stored as a learning value, and the fuel amount correction factor is multiplied by the current fuel amount for each cylinder to correct the fuel amount.

Description

Fuel amount learning and correction method and apparatus in gasoline direct injection engine {Method and apparatus for fuel amount training and compensation in GDI engine}

The present invention relates to a technology for monitoring an injection amount deviation of a gasoline direct injection engine cylinder, calculating and learning an oxygen sensor or engine roughness, and a fuel amount correction technology.

When starting a vehicle, the catalyst must be activated (above a certain temperature) to have the purification ability, and the faster the catalyst is activated, the more harmful exhaust gas can be reduced. To activate the catalyst, the engine ECU carries out a process of quickly raising heat toward the exhaust side (catalyst) through changes in valve timing and ignition timing. In the case of an engine equipped with a GDI (Gasoline Direct Injection) injector, the injector does not only inject fuel during the intake process, but also divides the injection before ignition.

This is called a catalyst heating section, and it is a section with poor combustion stability. In this section, the injection amount of the injector per injection is reduced because the injection is divided, and the injector injects a small flow rate (small flow rate).

This low-flow section is a section with a relatively large injection flow rate dispersion (deviation) of the injector compared to the medium and high-flow section. can make you feel uncomfortable.

Conventionally, cylinder deviation was diagnosed in order to correct the amount of fuel. To diagnose the cylinder deviation, the amount of fuel was forcibly changed and the deviation was learned with the roughness that occurred at that time. Therefore, a stable area in which the injection amount can be changed must be secured, and misfire may occur if the amount of fuel is greatly reduced during learning. In particular, it is difficult to learn the amount of fuel in the low-volume section of catalyst heating where roughness is unstable because misfires are highly likely to occur if the amount of fuel is modulated to learn the roughness deviation in the conventional method.

Accordingly, an object of the present invention is to solve the above-described conventional problems by correcting the amount of fuel in a low-volume section of catalyst heating injection where engine misfire due to roughness may occur most frequently.

In order to achieve the above object, the present invention improves fuel efficiency and performance of a vehicle by monitoring the deviation of each cylinder of the engine and correcting the fuel amount in response to the deviation. As the amount of fuel decreases in a specific driving area, the roughness of the oxygen sensor or engine is calculated at that time, and the deviation is learned, and the amount is corrected as the amount of fuel. In other words, the present invention calculates the average roughness of all cylinders and the difference between the roughness of each cylinder in the low flow rate section of catalyst heating, corrects the amount of fuel for the cylinders in which the roughness compared to the average is widened (exceeds a predetermined threshold value), and averages It converges to the roughness value.

More specifically, the present invention is a technology for preventing misfiring by correcting the roughness generated by the flow rate deviation of the injector in the low flow rate section of the catalyst heating injector (the injector has a greater deviation in the low flow rate section) for each cylinder. To this end, the lambda value (λ) (obtained through the test value) at which the roughness can return normally is calculated for the cylinder in which the roughness has occurred (exceeding a predetermined threshold).

In order to monitor the deviation between the cylinders, the average value of the cylinders of the roughness calculated from the current crank sensor is obtained, the difference between the average roughness and the individual cylinders is calculated, and the deviation is converted into a lambda value to calculate the fuel amount correction factor and learn it save as value Then, the fuel amount is corrected by multiplying the fuel amount correction factor by the current fuel amount for each cylinder.

The configuration and operation of the present invention will become more clear through specific embodiments described later in conjunction with the drawings.

Calculate the roughness difference of individual cylinders compared to the overall average cylinder, and calculate the roughness difference with the lambda value and the amount of fuel, and correct the fuel for the insufficient cylinder. There is no need to forcibly change the fuel to learn the deviation, and learning and fuel amount correction are possible even in the section where misfiring occurs.

Since the present invention is a method of calculating the difference in roughness between current cylinders without changing the amount of fuel in the catalyst heating low volume section and correcting the fuel of insufficient cylinders in real time, learning and correction are possible stably.

Through the present invention, combustion is stabilized in a low flow rate section of catalyst heating, and misfiring quality problems due to injector flow rate deviations can be prevented, failures caused by injectors can be prevented, and quality costs due to injector replacement can be reduced.

1 is a flow chart of a fuel amount learning and correction method according to the present invention
2 is actual engine roughness measurement data for verification of the fuel amount learning and correction method of the present invention
3 is a configuration diagram of a fuel amount learning and correction device according to the present invention
4 is a configuration diagram of another embodiment of a fuel amount learning and correcting device according to the present invention

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of preferred embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described below and may be implemented in various other forms. The examples are only provided to completely disclose the present invention and to completely inform those skilled in the art of the scope of the invention to which the present invention belongs, the present invention is defined by the description of the claims will be. In addition, terms used in this specification are for describing embodiments and are not intended to limit the present invention. In this specification, singular forms also include plural forms unless otherwise specified. Also, as used herein, the term "comprises" means the presence or absence of one or more other elements, steps, operations, and/or elements other than the mentioned elements, steps, operations, and/or elements; It is used in the sense of not excluding additions.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the embodiments, if a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

A misfiring of a vehicle is determined as a misfiring when the roughness of each cylinder reaches a certain value, and the roughness is calculated as a difference in crank angular velocity for each cylinder. In other words, if the cylinder is slower than other cylinders, the roughness rises, and this is judged as unstable combustion and misfire. Misfiring can occur when air volume/fuel amount calculation is incorrect or ignition fails, and it is particularly likely to occur when injector injection is not performed or fuel is insufficiently injected.

In the case of a misfire caused by fuel failure from the injector, the injector must be replaced because the injector is defective, but if the fuel is insufficient, misfire can be prevented by correcting the flow rate.

If the injection amount of a specific cylinder compared to all cylinders of the engine is less than the predetermined injection amount, a misfire may occur in that cylinder. In particular, in the multi-stage injection low volume section such as catalyst heating, the difference in flow rate percentage (%) may widen for each cylinder. Therefore, a device for stabilizing combustion through correction of fuel injection amount in this section is required.

Accordingly, the present invention is configured to perform fuel amount learning and correction for each cylinder (cylinder) only when catalyst heating (multi-stage injection) is performed at the initial stage of startup. That is, it is intended to reduce the deviation between the cylinders by correcting the deviation between the cylinders with fuel only in the catalyst heating section where the roughness is the weakest during the entire driving.

Under this background, the principle of fuel amount learning and correction according to the present invention will be briefly described. The fuel amount learning and correction method of the present invention is for correcting the fuel amount according to the injector deviation for combustion stability in the low flow rate (described above) section of a gasoline direct injection engine vehicle, 1) catalytic heating section through engine information and calculates the roughness of each cylinder and the average value of roughness through the crank angle sensor in the catalyst heating section. 2) Lambda (λ) deviation value for each cylinder corresponding to the deviation between average roughness and roughness for each cylinder ( In other words, a fuel amount correction factor is calculated through a corrected lambda value) and stored as a learning value. 3) The fuel amount correction factor is multiplied by the current fuel amount for each cylinder to correct the fuel amount.

Here, the correction lambda value means a value that can reduce the roughness of each cylinder that exceeds a predetermined threshold value compared to the average roughness, that is, a value that can reduce the roughness of each cylinder. Then the lambda value can be calculated.

In addition, the fuel amount correction factor is limited by the MAX value, and when the fuel amount correction factor does not reach the roughness improvement determination criterion even though the fuel amount correction factor reaches the MAX value, it is determined that the roughness deviation is not caused by an insufficient flow rate of the injector. , the fuel amount correction learning is stopped, and the factor is initialized.

Now, with reference to FIG. 1, the fuel amount learning and correction method 100 according to the present invention will be described in detail.

When the engine operates, it is first determined whether the catalyst heating condition is met (ie, the catalyst heating multi-stage injection situation is present) (110).

When it is determined that the catalytic heating is in progress, the average value of the engine roughness and roughness for each cylinder is calculated through the existing roughness calculation method (for example, calculated by the angular velocity of the crank angle sensor for each cylinder), and the roughness The roughness deviation (A) for each cylinder from the average varnish value is calculated (120).

A lambda (λ) deviation for each cylinder corresponding to each roughness deviation is calculated (130). The lambda deviation for each cylinder (in other words, the corrected lambda value) means a value that allows the roughness of each cylinder to converge to the average roughness value. In other words, if the roughness of a specific cylinder is large, it means that the roughness of other cylinders can be restored only when a lambda value corresponding to the roughness is applied to the corresponding cylinder. The lambda value can be obtained using a linear relationship (test value). As an example, the lambda value is calculated by entering the roughness deviation into the value made by the two-dimensional curve obtained from the test value.

A fuel compensation factor is calculated by integrating the lambda deviation for each cylinder (140). A fuel compensation factor (C) for each cylinder is calculated (150), and an average value (B) of all cylinders of the fuel compensation factor is obtained (160).

The fuel amount correction factor is obtained by calculating the ratio of the fuel compensation factor for each cylinder to the average fuel compensation factor for all cylinders. To this end, the value obtained by adding 1 to the fuel compensation factor for each cylinder is divided by the value obtained by adding 1 to the average fuel compensation factor for all cylinders (ie, (1+C)/(1+B)) (170). A fuel amount correction factor is applied to a cylinder whose roughness occurs due to lack of fuel, so that the injection time is further increased, and the roughness of the cylinder is reduced to prevent misfire.

The fuel amount correction factor is stored as a learning value because it must be used right away at the next startup as well as calculation (180).

The fuel amount is corrected by calculating a new fuel amount for each cylinder by multiplying the current fuel amount for each cylinder by the fuel amount correction factor (190).

The fuel amount correction factor is limited by the MAX value determined for it. If the roughness does not return (i.e., does not improve) even when the fuel amount is corrected, the correction factor may continue to increase, and a device for preventing this is required. To this end, it is determined whether the fuel amount correction factor has reached MAX and whether the roughness has reached a set value ('roughness improvement determination reference value') determined to be improved (200). As a result of the determination, even though the fuel amount correction factor has reached MAX, if the difference between the A value at the time of initial learning (the deviation of average roughness for each cylinder) and the current A value does not reach the roughness improvement judgment reference value (205), the amount of fuel The correction learning is stopped and the fuel amount correction factor is initialized (210). This is because there is a high possibility that it is not roughness caused by the injector flow rate deviation, and thus the fuel correction amount is meaningless. Here, the MAX value of the fuel amount correction factor and the roughness improvement judgment reference value are determined as test values.

Meanwhile, when the catalyst heating is finished, the fuel amount correction is stopped, and the fuel amount correction factor is stored and used as an initial value when catalyst heating is performed next.

2 shows actual engine roughness measurement data for verification of the fuel amount learning and correction method of the present invention described above. This is the result obtained by creating a situation where the fuel amount of the 4th cylinder is 20% insufficient, forcibly starting the catalytic heating, and performing the fuel amount correction logic.

In (a) of FIG. 2, K denotes a forced catalyst heating section at startup, L denotes a misfire occurrence section, and M denotes a misfire removed section. In the forced catalytic heating section (K), the amount of fuel in cylinder 4 was 20% less than that in cylinders 1 to 3, so a misfire occurred due to engine roughness (10) (L) (roughness (10) exceeded the threshold value Thd). If it exceeds, it is determined as a misfire), and as a result of increasing the fuel amount learning value (20) of the 4th cylinder according to the present invention, the fuel amount of the 4th cylinder is increased, the roughness of the 4th cylinder is stabilized (M), and the misfire disappears. can see.

(b) of FIG. 2 shows the engine roughness 10' when the catalyst heating is restarted after the engine is turned off. It can be confirmed that the engine roughness 10' is quickly stabilized because the stored fuel amount learning value 20 is loaded and used as the initial fuel amount value 20' for injector injection.

3 is a configuration diagram of a fuel amount learning and correcting device 10 according to the present invention. It represents a functional hardware (and/or software) configuration for realizing the fuel amount learning and correction method 100 described above.

As shown in FIG. 3 , the fuel amount learning and correcting device 10 calculates the difference between the average value of the individual cylinder roughness and the individual cylinder roughness of the engine 5, and diagnoses the roughness deviation of each cylinder. varnish diagnosis unit 20; a lambda calculation unit 30 that calculates a correction lambda value capable of converging the roughness of each cylinder to an average roughness value; and a control unit 40 that calculates a fuel amount correction factor by integrating the correction lambda values and compensates by multiplying the amount by the existing fuel amount.

Here, the roughness diagnosing unit 20 is configured to acquire the roughness by the angular velocity of the crank angle sensor for each cylinder when calculating the roughness for each cylinder.

In addition, the correction lambda value calculated by the lambda calculation unit 30 means a value that can reduce the roughness of each cylinder widened compared to the average roughness, and the roughness deviation to the value made by the two-dimensional CURVE obtained from the test value If entered, the lambda value can be calculated.

The control unit 40 generates a fuel compensation factor by integrating correction lambda to calculate the fuel amount correction factor; Calculate the fuel compensation factor for each cylinder, and calculate the average value of the fuel compensation factor for all cylinders; Calculate the fuel amount correction factor by dividing the value obtained by adding 1 to the fuel compensation factor for each cylinder by the value obtained by adding 1 to the average fuel compensation factor for all cylinders; The calculated fuel amount correction factor is stored as a learning value, and compensation is performed by multiplying the current fuel amount for each cylinder by the fuel amount correction factor.

4 is a configuration diagram of another embodiment of a fuel amount learning and correcting device 10 according to the present invention.

A limiting unit 50 is added to the configuration of FIG. 3 . The limiting unit 50 determines that the fuel amount correction factor calculated by the control unit 40 is limited by a preset MAX value, and even if the fuel amount correction factor reaches the MAX value, the roughness improvement judgment standard value is not reached. In this case, it is determined that the roughness deviation is not caused by insufficient flow rate of the injector, and the fuel amount correction learning is stopped and the factor is initialized.

The limiting unit 50 for stopping the fuel amount correction learning and initializing the factor determines the deviation value of individual cylinders from the average engine roughness of each cylinder at the time of initial learning. Calculate a value minus the deviation value; determining whether the roughness is not improved by determining whether the above value is smaller than a roughness improvement judgment value; judging whether the fuel amount correction factor has reached the Max value; When both the correction factor MAX judgment and the roughness non-improvement judgment are satisfied, the fuel amount correction learning is stopped and the fuel amount correction factor is initialized.

Here, the MAX value of the fuel amount correction factor and the reference value for determining roughness improvement may be determined by test values.

In the above, the present invention can be implemented in terms of apparatus or method. In particular, the functions or processes of each component of the present invention are digital signal processors (DSPs), processors, controllers, and application-specific ASICs. IC), a programmable logic device (FPGA, etc.), at least one of other electronic devices, and a hardware element including a combination thereof. In addition, it can be implemented as software combined with hardware elements or independently, and this software can be stored in a recording medium.

So far, the present invention has been described in detail through preferred embodiments of the present invention, but those skilled in the art to which the present invention pertains may differ from the contents disclosed herein without changing the technical spirit or essential features of the present invention. It will be appreciated that it may be embodied in other specific forms. It should be understood that the embodiments described above are illustrative in all respects and not restrictive. In addition, the scope of protection of the present invention is determined by the claims described below rather than the detailed description above, and all changes or modifications derived from the scope of the claims and equivalent concepts should be construed as being included in the technical scope of the present invention. do.

Claims (13)

delete delete delete When the engine is running, it is determined whether the catalytic heating multi-stage injection situation is present;
If it is determined that the catalyst is heating, engine roughness and an average value of roughness for each cylinder are calculated, and a deviation of roughness for each cylinder (hereinafter, referred to as 'A value') from the average value of roughness is calculated;
Calculate a lambda deviation for each cylinder corresponding to the deviation of each roughness, wherein the lambda deviation for each cylinder means a value that allows the roughness of each cylinder to converge to an average roughness value;
calculating the fuel compensation factor for each cylinder by integrating the lambda deviations for each cylinder, and obtaining an average value of the fuel compensation factor for all cylinders;
obtaining a fuel amount correction factor by calculating the ratio of the fuel compensation factor for each cylinder to the average fuel compensation factor for all cylinders;
storing the fuel amount correction factor as a learning value, calculating a new fuel amount for each cylinder by multiplying the current fuel amount for each cylinder by the fuel amount correction factor, and correcting the fuel amount;
determining whether the fuel amount correction factor has reached a preset MAX value and a set value at which it is determined that the roughness is improved (hereinafter referred to as 'roughness improvement judgment value');
As a result of the determination, even though the fuel amount correction factor has reached the MAX value, if the difference between the A value at the initial learning and the current A value does not reach the roughness improvement judgment reference value, the fuel amount correction learning is stopped and the fuel amount correction factor is initialized. A fuel amount learning and correction method in a gasoline direct injection engine comprising doing. The method of claim 4, wherein the lambda deviation, the MAX value of the fuel amount correction factor, and the roughness improvement determination reference value are calculated using test values. The method of claim 4, in order to obtain a fuel amount correction factor by calculating the ratio of the fuel compensation factor for each cylinder to the average fuel compensation factor for all cylinders,
A fuel amount learning and correction method in a gasoline direct injection engine comprising dividing a value obtained by adding 1 to a fuel compensation factor for each cylinder by a value obtained by adding 1 to an average fuel compensation factor for all cylinders. a roughness diagnosis unit for diagnosing deviations in roughness of each cylinder by calculating an average value of roughness of individual cylinders of the engine and a difference value between roughnesses of individual cylinders;
A correction lambda value that allows the roughness of each cylinder to converge to the average roughness value - Here, the correction lambda value means a value that can reduce the roughness of each cylinder that exceeds a predetermined threshold compared to the average roughness - A lambda calculation unit that calculates; and
Including a control unit that calculates a fuel amount correction factor by integrating the correction lambda value and compensates by multiplying the existing fuel amount,
The control unit
To calculate the fuel amount correction factor, a correction lambda is integrated to generate a fuel correction factor; calculating a fuel compensation factor for each cylinder, and calculating an average value of the fuel compensation factor for all cylinders; Calculate a fuel amount correction factor by dividing the value obtained by adding 1 to the fuel compensation factor for each cylinder by the value obtained by adding 1 to the average fuel compensation factor for all cylinders; A fuel amount learning and correcting device in a gasoline direct injection engine configured to store the calculated fuel amount correction factor as a learning value and compensate by multiplying the current fuel amount for each cylinder by the fuel amount correction factor. 8 . The device of claim 7 , wherein the roughness diagnosis unit acquires the roughness by the angular velocity of the crank angle sensor for each cylinder when calculating the roughness for each cylinder. The device of claim 7, wherein the corrected lambda value calculated by the lambda calculation unit is calculated by inputting a roughness deviation into a value made of a two-dimensional curve obtained as a test value. delete The method of claim 7, wherein the fuel amount correction factor calculated by the control unit is limited by a preset MAX value, and when the fuel amount correction factor does not reach the roughness improvement determination criterion value even though the fuel amount correction factor reaches the MAX value, the injector flow rate An apparatus for learning and correcting fuel amount in a gasoline direct injection engine, further comprising a limiting unit that determines that the roughness deviation is not caused by the shortage and stops fuel amount correction learning and initializes the factor. The method of claim 11, wherein the limiting unit
Calculate a value obtained by subtracting the current deviation value of each cylinder from the average engine roughness per cylinder from the deviation value of each cylinder from the average engine roughness per cylinder at the time of initial learning; determining whether the roughness is not improved by determining whether the above value is smaller than the roughness improvement judgment value; determining whether the fuel amount correction factor has reached the Max value; and a fuel amount learning and correcting device in a gasoline direct injection engine configured to stop fuel amount correction learning and initialize a fuel amount correction factor when both the correction factor MAX judgment and the roughness non-improvement judgment are satisfied. 12. The device of claim 11, wherein the fuel amount correction factor MAX value and the roughness improvement determination reference value are determined as test values.

Images