JPH07139393A - Fuel injection control device for internal combustion engine - Google Patents

Abstract

PURPOSE:To provide a fuel injection control device for an internal combustion engine which is constituted to improve calculation precision of a fuel adhesion model parameter and perform high-precise adaptive control of a fuel injection amount. CONSTITUTION:A fuel injection control device for an internal combustion engine 10 contains a parameter setting means 11 to calculate, estimate, and set a fuel behavior model parameter based on detecting data indicating the state of the internal combustion engine 10 and a fuel injection control means 12 to perform adaptive control of a fuel injection amount by using a set parameter. A parameter setting means 11 is operated to detect a transition period wherein a change amount of an intake air amount exceeding a given level is continued throughout a plurality of cycles of calculation is detected and calculate and estimate a parameter only during transition. The parameter setting means, simultaneously, is operated to update a parameter in a way that weighing is effected according to the intensity of transition during transition to average parameters calculated in the past. Further, the means is worked to monitor a relation between the parameter and the operation state of the internal combustion engine and prohibit update processing of a parameter when abnormality of the updating content of a calculated parameter is detected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection control device for an internal combustion engine, and more particularly to a fuel injection control device for an internal combustion engine that automatically sets a parameter indicating a behavior characteristic of fuel to adaptively control a fuel injection amount.

[0002]

2. Description of the Related Art In fuel injection control of an internal combustion engine, based on a fuel behavior model in an intake pipe, an adhesion ratio indicating a ratio of injected fuel adhering to a wall surface of an intake pipe and a ratio of evaporation of fuel adhering to a wall surface of the intake pipe are shown. A technique for estimating and calculating an evaporation rate (or a carry-out rate indicating a rate at which adhered fuel is carried away to a cylinder) by a least-squares method, and optimally controlling a fuel injection amount by using those fuel behavior model parameters is disclosed in, for example, Japanese Patent Laid-Open No. Hei 10-242, It is known from JP-A-2-173334.

FIG. 5 is an explanatory view showing an example of a fuel behavior model of inflow, adhesion, and evaporation of fuel into the intake pipe. Injected fuel f _i , and adhering fuel f remaining adhering to the pipe wall and the like.
_{Assuming that w} and the combustion fuel f _c that flows out and is used for combustion in the cylinder have the relationship shown in the figure by the model parameter fuel attachment rate r and the fuel residual rate (= 1−evaporation rate) p, It is to be modeled. That is, in each calculation cycle (k cycle), _{fw (k + 1)} = pfw _(k) + rfi _(k) fc _(k) = (1-p) _{fw (k)} + (1- r) f _{i (k)} is established, f _i and f _c that can be calculated from the actual measurement data are used, and the model parameters r and p are calculated by the following equations and from three calculation cycles of, for example, k, k + 1, and k + 2. You can ask.

P (f _{c (k)} -f _{i (k)} ) + r (f _{i (k)} -f _{i (k + 1)} ) + f _{i (k + 1)} -f _{c (k + 1)} = 0 p (f _{c (k + 1)} -f _{i (k + 1)} ) + r (f _{i (k + 1)} -f _{i (k + 2)} ) + f _{i (k + 2)} -f _{c (k + 2)} = 0 The solution of the binary linear simultaneous equations can be obtained by using, for example, the least square method.

[0005]

However, when the model parameters r and p are calculated from the actual data of f _i and f _c , there is insufficient precision (measurement error) or noise in each detected data, and There are cases where the effect of the error is large and cases where the case is not large, and there are large variations in each calculation, and accurate values of r and p cannot be obtained. Also,
A huge amount of calculation is required to estimate and calculate the parameters by the recursive least squares method, and there is a problem that the convergence is slow in order to suppress the variation by averaging.

On the other hand, the influence of the adhered fuel does not appear in the steady operation but appears in the transient operation. According to the above and equations, both equations are the same during steady operation, r,
I can't solve p. Further, even if the calculation is performed during the transient operation, since the data of the three calculation cycles k, k + 1, and k + 2 are used, there is a problem that the transient operation needs to be continued for three calculation cycles or more.

Therefore, an object of the present invention is to provide a fuel injection control device for an internal combustion engine which improves the calculation accuracy of the fuel adhesion model parameter and enables highly accurate adaptive control of the fuel injection amount. Further, the present invention provides a fuel injection control device for an internal combustion engine that enhances the calculation accuracy of the fuel adhesion model parameter by weighting a specific calculation condition and a calculation result, and enables highly accurate adaptive control of the fuel injection amount. The purpose is to

Further, the present invention improves the parameter calculation accuracy by ignoring the abnormal calculation result of the fuel adhesion model parameter, and makes it possible to perform highly accurate adaptive control of the fuel injection amount. The purpose is to provide.

[0009]

In a fuel injection control device for an internal combustion engine according to the present invention, a parameter indicating a behavior characteristic of fuel including adhesion of injected fuel to an intake pipe wall surface is based on detection data indicating a state of the internal combustion engine. In the fuel injection control device for the internal combustion engine, which includes the parameter setting means for calculating and estimating and setting, and adaptively controlling the fuel injection amount using the parameter updated by the parameter setting means, the parameter setting means has a predetermined level or more. Detects a transient time when the amount of change in the intake air amount continues for multiple cycles of the above calculation,
The parameters are updated by calculating and estimating the parameters during the transition, and weighting according to the intensity of the transition during the transition and averaging the parameters with those calculated in the past.

Further, there is provided parameter setting means for calculating and estimating a parameter indicating a behavior characteristic of the fuel including adhesion of the injected fuel to the wall surface of the intake pipe based on detection data indicating a state of the internal combustion engine, and setting the parameter. In the fuel injection control device for the internal combustion engine that adaptively controls the fuel injection amount using the parameter updated by the means, the parameter setting means monitors the relationship between the parameter and the operating state of the internal combustion engine, and calculates the above. When an abnormality is detected in the parameter update content, the parameter update process is prohibited.

[0011]

According to the above configuration, the calculation condition that the transient time is continuing is added to the calculation of the model parameter,
Moreover, by highly reflecting the highly reliable calculation result corresponding to the strength of the transition, it is possible to calculate and estimate and set the highly accurate model parameter. In addition, it is possible to more accurately calculate and estimate the model parameters by detecting the calculation results that are clearly wrong from the relationship between the model parameters and the operating state of the engine, and prohibiting the parameter updating process with such values. Becomes

[0012]

1 is a block diagram showing the basic construction of a fuel injection control system for an internal combustion engine to which the present invention is applied. Here, based on the fuel behavior model in the intake pipe, the fuel adhesion rate r and the residual rate p of the adhered fuel p, that is, (1-evaporation rate) are estimated and calculated, and the fuel behavior model parameters r and p are used to calculate the fuel. The injection amount is optimally controlled. 1 in the figure
Reference numeral 0 is an internal combustion engine, 11 is a parameter setting means for estimating and calculating a fuel behavior model parameter using various detected amounts of the internal combustion engine 10 and detected values of air-fuel ratio in exhaust gas, 1
Reference numeral 2 denotes a fuel injection control unit that uses the parameter from the parameter setting unit 11 to control the fuel injection amount so that the internal combustion engine 10 has the target air-fuel ratio.

FIG. 2 is a schematic configuration diagram for explaining a hardware configuration of a fuel injection control device for an internal combustion engine to which the present invention is applied. Here, the throttle sensor 1 for detecting the state of the internal combustion engine 10, the intake pressure sensor 2, the fuel injection injector 3, the air-fuel ratio sensor 4, the exhaust gas purification catalyst 5, and the state of the internal combustion engine 10 are shown in FIG. The control unit 20 for executing the operation control of the internal combustion engine 10 including the parameter setting means 11 for controlling the fuel injection amount and the fuel injection control means 12 and the like is shown.

In the parameter setting means 11 for controlling the fuel injection amount in FIG. 1 described above, for example, various state detection amounts and exhaust gas of the internal combustion engine 10 that are input at a predetermined crank angle position of the internal combustion engine or at regular intervals. The data such as the air-fuel ratio detection value in the inside is taken in, and the calculation formulas of the above model parameters formed from the data of the past three cycles and the above are solved, and the model parameters r and p are calculated and estimated. It is set.

FIG. 3 is a flow chart showing an example of software for realizing the first embodiment of the fuel injection control apparatus for the internal combustion engine according to the present invention. In the present embodiment, the condition for starting the calculation is that the calculation cycle has continued for three cycles or more, and the two-dimensional linear simultaneous equations are solved by using the inverse matrix and the reliability of the calculation result is determined by the determinant. , The weighting is applied to the calculation result, and the accuracy of the model parameters r and p is improved.

[0016] Therefore, firstly, and in step 301, the cylinder intake air quantity PMFWD detection data, by using the fuel injection amount TAU and air-fuel ratio A / F, the injected fuel i.e. injector fuel injection amount f _i from TAU,, combustion The fuel, that is, the fuel inflow amount in the cylinder f _c is PM
Calculate from FWD / (A / F). This calculation is executed for three cycles, and f _{i (k)} , f _{i (k + 1)} , f _{i (k + 2)} , f _{c (k)} , f
Collect each data of _{c (k + 1)} and fc _{(k + 2)} . Next, in step 302, it is determined whether the collected data for three cycles is for transient or for steady state. Specifically, it is judged by whether the amount of change in the intake air amount above a specified value has continued for three cycles, and
(Yes), that is, "PMFWD _(k) and PMFWD
_When the difference from _{(k + 1)} is greater than or equal to a predetermined value and the difference between PMFWD _{(k + 1)} and PMFWD _{(k + 2)} is greater than or equal to the predetermined value, it is determined that the transition time is present. When the answer is No, that is, in the steady state, this calculation routine is terminated and not executed.

If it is determined that the transition is in progress, step 3
At 03, the f _i collected at step 301,
Substituting each data of f _c into the above and the equations, the obtained binary binary simultaneous equations are solved by calculating the inverse matrix. In this case, the size of the determinant indicates the strength of the transient state, and is mathematically related to the intersection of the intersections of the two straight lines indicating the solution of the above two-dimensional simultaneous linear equation, The larger the size, the closer the intersection of the two straight lines becomes to a right angle. Therefore, if the determinant is small, the two straight lines are close to parallel, and even if there is a slight error in the measurement data, the error in the calculation result will be large. Therefore, the size of the determinant is taken as the reliability of the calculation result, and the weighted average of the calculation result is performed using it. Therefore, step 30
In 3, the model parameters r and p are calculated by the inverse matrix calculation, and the size of the determinant at that time is calculated.

Next, when this work is repeated a plurality of times,
Although r and p are calculated with variations, they are weighted by the size of the determinant when they are averaged. That is, in step 304, r and p are calculated by the following equation.

[0019]

[Equation 1]

According to the first embodiment, the weighting for the calculation result is increased during the period when the error included in the detection data by the sensor detection or the like is relatively small, that is, during the transition, and the smoothing process is performed. By doing so, it is possible to reduce the influence on the error. That is, by adding the calculation condition to the calculation of the model parameter and reflecting the calculation result with high reliability largely, it is possible to calculate and set the model parameter with high accuracy. Further, since the calculation is executed only under a specific condition, the calculation load can be reduced.

FIG. 4 is a flow chart showing an example of software for realizing the second embodiment of the fuel injection control apparatus for the internal combustion engine according to the present invention. In the present embodiment, when the model parameters are calculated, a clearly wrong calculation result, for example, when the model parameters r and p are both too large or too small in consideration of the operating state (vehicle speed and air-fuel ratio), is ignored. By doing so, the accuracy of calculation and estimation of the parameters is improved.

For this reason, first, at step 401, the currently known r, p, that is, r ₀ , p ₀ is used to perform the adhesion correction of the fuel injection amount. Specifically, the basic formula of the fuel adhesion model described above with reference to FIG. 5 is fw _{(k + 1)} = pfw _(k) + rfi _(k) fc _(k) = (1-p) f based on the _{w (k) + (1-} r) f i (k), f i (k) = _{[f cr (k) - (1} -p) f w (k) ] / (1-r) f _{w (k + 1)} = pf _{w (k)} + rf _{i (k)} Here, f _cr calculates the fuel amount required for combustion calculated from the cylinder intake air amount, and controls the fuel injection amount.

Next, at step 402, it is judged whether the air-fuel ratio is in a rich or lean state above a predetermined level, and whether it is in an acceleration or deceleration state above a predetermined level. Here, the determination of the acceleration or deceleration state equal to or higher than the predetermined level is, for example, the acceleration state if the change amount of the cylinder intake air amount PMFWD is equal to or more than the first predetermined value, and the acceleration state is equal to or less than the second predetermined value. This can be realized by setting the deceleration state.

In the judgment of step 402, the air-fuel ratio is in a rich or lean state above a predetermined level,
If it is not in the acceleration or deceleration state above the predetermined level (No), this routine is terminated and the updating process of the model parameters r and p is prohibited, but if both of the states are established (Yes). Advances to step 403, and as described above, the binary linear simultaneous equations and are obtained from f _i and f _c for three cycles, and they are solved, model parameters r and p are calculated, and r ₁ , p _{Get one} .

Here, the model parameters r ₁ and p ₁ obtained by the calculation are monitored by comparing them with the model parameters r ₀ and p ₀ actually used for the control while considering the operating state of the internal combustion engine. If the relationship is abnormal, it is determined that the calculation is incorrect and the model parameters r and p are not updated. For example, when the fuel injection control is executed using the above-described fuel behavior model shown in FIG. 5, there is a tendency as shown in the following table.

P too large, r too large p too small, r too small ───────────────────────────── Acceleration Rich lean ─────── ────────────────────── Deceleration Lean Rich ────────────────────────── ───

The software of FIG. 4 is an example using this relationship. If the operating state is acceleration rich or deceleration lean, the model parameters r ₀ , p actually used for control are used.
₀ means that at least one of them is excessive, but nevertheless p ₁ > p ₀ and r ₁ > r ₀
If so, the model parameters r and p are not updated and the model parameters r and p are not updated (step 404). If the operating state is acceleration lean or deceleration rich, the model parameter r actually used for control is calculated. _At least one of ₀ and p ₀ is too small. However, if p ₁ <p ₀ and r ₁ <r _0, it is determined that this is also a miscalculation. , The model parameters r and p are not updated (step 405). Then, when the determination at step 404 or step 405 is (No), it is determined that the model parameters r ₁ and p ₁ obtained by calculation have been correctly calculated, and the model parameters r ₀ and p ₀ actually used for control are set. Update based on r ₁ and p ₁ respectively (step 406).

According to the present embodiment, it is possible to detect a clearly wrong calculation result, for example, when a wrong parameter is calculated due to a sensor abnormality or the like, and the parameter updating process with such a value is prohibited. , More accurate calculation and estimation of model parameters becomes possible. Further, since the parameter erroneous learning is eliminated, the parameter learning and updating speed can be improved.

[0029]

As described above, according to the fuel injection control device for an internal combustion engine of the present invention, by adding a calculation condition to the calculation of the model parameter and reflecting a highly reliable calculation result largely, It is possible to reduce the computational load while calculating and setting highly accurate model parameters. In addition, it is possible to detect a wrong calculation result clearly, prohibit the parameter updating process with such a value, and calculate and estimate the model parameter more accurately, and improve the learning update speed of the parameter. You can Therefore, according to the present invention, it is possible to improve the controllability of the fuel injection control device of the internal combustion engine that adaptively controls the fuel injection amount by automatically setting the parameter indicating the fuel behavior characteristic.

[Brief description of drawings]

FIG. 1 is a block diagram showing a basic configuration of a fuel injection control device for an internal combustion engine to which the present invention is applied.

FIG. 2 is a schematic configuration diagram for explaining a hardware configuration of a fuel injection control device for an internal combustion engine to which the present invention is applied.

FIG. 3 is a flowchart showing an example of software for realizing the first embodiment of the fuel injection control device for the internal combustion engine according to the present invention.

FIG. 4 is a flowchart showing an example of software for realizing a second embodiment of the fuel injection control device for the internal combustion engine according to the present invention.

FIG. 5 is an explanatory diagram showing an example of a fuel behavior model.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Throttle sensor 2 ... Intake pressure sensor 3 ... Fuel injection injector 4 ... Air-fuel ratio sensor 5 ... Exhaust gas purification catalyst 10 ... Internal combustion engine 11 ... Parameter setting means 12 ... Fuel injection control means 20 ... Control unit

Claims (2)

[Claims] 1. A parameter setting means for calculating and estimating a parameter indicating a behavior characteristic of fuel including adhesion of injected fuel to a wall surface of an intake pipe based on detection data indicating a state of an internal combustion engine, and setting the parameter. In the fuel injection control device for an internal combustion engine that adaptively controls the fuel injection amount using the parameter updated by the means, the parameter setting means causes the change amount of the intake air amount of a predetermined level or more to continue for a plurality of cycles of the calculation. The parameter is updated by detecting the transient time, calculating and estimating the above parameter during the transient time, weighting according to the strength of the transient at the time of the transient, and averaging with the parameter calculated in the past. A fuel injection control device for an internal combustion engine, comprising: 2. A parameter setting means for calculating and estimating a parameter indicating a behavior characteristic of fuel including adhesion of injected fuel to a wall surface of an intake pipe based on detection data indicating a state of an internal combustion engine, and setting the parameter. In the fuel injection control device for the internal combustion engine that adaptively controls the fuel injection amount using the parameter updated by the means, the parameter setting means monitors the relationship between the parameter and the operating state of the internal combustion engine, and calculates the above. A fuel injection control device for an internal combustion engine, wherein the parameter update process is prohibited when an abnormality is detected in the parameter update content.