JP2997473B2 - Engine adaptive control method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fuel injection control device for an automobile engine, and more particularly to an adaptive control method for an engine suitable for automatically setting control parameters thereof.

[Conventional technology]

Conventional devices are disclosed in Japanese Patent Application Laid-Open Nos. 58-8238, 60-201042, 60-
The adhesion rate, which is the rate at which the injected fuel adheres to the intake pipe wall surface, and the evaporation rate, which is the rate at which the fuel adhered to the intake pipe wall evaporates per unit time, as described in No. 126337, The correspondence between the carry-out rate, which is the carry-out rate, and the various detected amounts of the engine such as the intake air amount and the intake pipe internal pressure is determined in advance, and the fuel amount to be supplied to the cylinder is determined using the correspondence. I have. That is, using the correspondence obtained in advance, the control parameters such as the adhesion rate and the evaporation rate (or,
The carry-out rate is calculated, and the fuel injection amount is determined based on the calculation result.

[Problems to be solved by the invention]

The prior art described above has a problem in that parameters such as an adhesion rate and an evaporation rate need to be calculated in advance in various operation regions by an engine test, and a man-hour is required for control system development.

In addition, it is necessary to match the calculation parameters for each engine due to individual differences in engine characteristics, which also requires a man-hour.

In addition, there is a problem that the setting parameter is deviated from the optimum value with the passage of time and controllability is deteriorated with no consideration given to the change over time of the engine.

Further, in the conventional method, no consideration is given to the influence of the delay of the measurement of the air amount and the delay of the transportation. These delays cause deterioration in the controllability during the transition. It is conceivable that this delay can be collected into parameters such as the adhesion rate and the evaporation rate, that is, the desired controllability can be obtained in an area where parameter matching is performed. However, there has been a problem that desired controllability cannot be obtained in various areas. In other words, the characteristics of the air amount measurement delay are completely different between acceleration and deceleration, so even if parameter matching is performed to obtain the desired controllability during acceleration, controllability will deteriorate during deceleration or during deceleration. Even if parameter matching is performed, there is a problem that controllability is deteriorated during acceleration.

A first object of the present invention is to provide an adaptive control method for an engine that automatically sets and corrects the above-mentioned parameters in accordance with the variation over time of an individual difference in engine characteristics.

It is a second object of the present invention to provide an adaptive control method of an engine that compensates for the influence of a delay in the amount of air and obtains desired controllability in various operating regions.

[Means for solving the problem]

The first object is to determine the adhesion rate, which is the rate at which the injected fuel adheres to the intake pipe wall surface from various detected amounts of the engine, and the evaporation rate, which is the rate at which the fuel adhered to the intake pipe wall evaporates per unit time (or An engine control device that calculates a carry-out rate, which is a rate at which fuel is carried out to a cylinder per unit time, and controls the amount of fuel supplied to the cylinder based on the calculated value. A wide-range air-fuel ratio sensor is installed in an exhaust pipe. Parameter estimation means for estimating the adhesion rate, the evaporation rate, or the carry-out rate from the air-fuel ratio measured by the sensor and the fuel injection amount measured air amount, etc .; This is achieved by providing parameter characteristic correction means for correcting the correspondence between the rate and the evaporation rate or the carry-out rate.

The first and second objects are to install a wide-range air-fuel ratio sensor in the exhaust pipe, and determine the adhesion rate, the evaporation rate,
Alternatively, a parameter estimating means for estimating the carry-out rate is provided, and instead of the adhesion rate, the evaporation rate, and the carry-out rate calculated from the various detected amounts, the fuel quantity supplied to the cylinder is controlled by the estimating means based on the estimation result. Is achieved by

Further, the second object is achieved by providing a plurality of types of correspondences between the various detection amounts and the adhesion rate, the evaporation rate, and the carry-out rate according to the driving situation.

[Action]

In the means of FIG. 1, the parameter estimating means estimates a model parameter, such as an adhesion rate, an evaporation rate, or a removal rate, based on a model representing a fuel flow characteristic in the intake pipe, for example, by a method of sequential least squares. This is sent to the parameter characteristic correcting means in combination with the measured value of the engine operating state at the time of estimation. The parameter characteristic correction means corrects the correspondence between the various detected amounts and the adhesion rate, the evaporation rate, and the carry-out rate of the engine, which are appropriately determined in advance based on the obtained data, and the operation state at each time. Define the optimal relationship in. As described above, even if the initial settings are appropriate, as the number of engine operations increases, the parameters such as the adhesion rate and the evaporation rate are automatically optimized, so that the number of man-hours required for system development is reduced, and individual differences in engine characteristics over time. Adaptation to is possible.

In the second means, the same effect can be obtained because the adhesion rate, the evaporation rate, and the carry-out rate are always kept at the optimum values in the occasional operation state by the parameter estimation means.

In the third means, different parameters can be set according to the operation state such as acceleration and deceleration, so that desired controllability can be obtained in various operation regions.

〔Example〕

An embodiment of the present invention will be described below with reference to FIGS.

FIG. 1 is an overall configuration diagram of the first means of the present invention. In contrast to the conventional device comprising only the parameter calculation means and the fuel injection control means, the apparatus is newly provided with a parameter estimation means and a parameter characteristic correction means.

 Hereinafter, the operation of each means will be described.

The parameter calculation means calculates the rate of attachment of the injected fuel to the wall of the intake pipe from the measured values of the operating state of the engine, such as the amount of intake air and water temperature, and the rate of evaporation of the fuel attached to the intake pipe per unit time. Is calculated on the basis of a predetermined correspondence relationship, and this is sent to the fuel injection control means. The fuel injection control means controls the amount of fuel supplied to the cylinder based on the calculated adhesion rate and evaporation rate.

The parameter estimating means is a control parameter of an adhesion rate, an evaporation rate (or a carry-out rate) from time-series data such as an exhaust gas air-fuel ratio and a fuel injection amount measured air amount measured by a wide area air-fuel ratio sensor installed in an exhaust pipe. Is estimated. The parameter estimation can be performed, for example, by applying the sequential least squares method to a discrete equation of the following mathematical model representing the fuel flow characteristic in the intake pipe.

Here, G _f : fuel injection amount (g / s) (converted value of injected fuel amount into mass flow per unit time) G _fe : cylinder inflow fuel amount (g / s) M _f : intake pipe wall adhering fuel amount (G) X: Adhesion rate 1 / τ: Evaporation rate (1 / s) (or carry-out rate) X, 1 / τ is regarded as a constant, and equations (1) and (2) are discretized and G _fe
Deriving the relational expression between Gf and Gf gives the following expression. The central difference is used for the continuous difference calculation.

In the equation (3), since the error equation is linear with respect to the parameter X, 1 / τ, it is possible to calculate an optimal parameter in the sense that the weighted sum of squares of the equation error is minimum.

That is, even if the evaluation index is set to equation (4), X, 1 / τ that minimizes J can be calculated, and the recursive algorithm is given by equations (5) and (6).

As a recursive parameter estimation method, besides the equations (5) and (6), the theory and practice of an adaptive control system (Ohm)
The method described on pages 78 to 86 can also be applied.

According to the equations (5) and (6),
In order to estimate and calculate X, 1 / τ, data of G _f (k) and G _fe (k) is required.

The fuel injection amount G _f (k) is calculated from a command value for the fuel injection valve of the microcomputer. Fuel flow into cylinder G _fe (k)
Since there is no sensor for directly measuring this, it is calculated by the following equation from the measured values in other operating states.

Here Q _a, in the L di Etro Nix system and those subjected noise, various processes for pulsating retirement the measurement value of the air quantity Surotsutoru upstream detected by the air flow sensor. That is, the air amount is used as a base for determining the fuel injection amount. Further, in the D dietronics system, the amount of air flowing into the cylinder is calculated based on the intake pipe internal pressure detected by the pressure sensor. This is also the air amount that is the basis for determining the fuel injection amount.

A / F is an exhaust gas air-fuel ratio detected by an air-fuel ratio sensor installed in an exhaust pipe.

ko is the average time from when a mixture of air and fuel is drawn into the cylinder to when the combustion gas is exhausted.
Here, it is assumed that it corresponds to three strokes of the piston movement (11)
Formulate by formula.

In order to calculate the cylinder inflow fuel amount in (10), originally air amount Q _a is should be mass air flow into cylinder. However, in the L-jetronics system, a measured value of the air passing through the throttle is used. Further, even in the D-jetronics system, the measured value cannot match the true cylinder inflow air amount due to a response delay of the sensor. In (10), a Q _a not a true cylinder intake air quantity, to use the underlying air quantity of fuel injection quantity determined in each system is the following reason. In the conventional method, the desired controllability is obtained by compensating only for the fuel delay.However, in actuality, there is also a delay in air transportation and measurement, which is a deterioration in controllability during rapid acceleration / deceleration. Cause The reason for using the amount of air, which is the basis for fuel determination, to consolidate the air delay to some extent X, 1 / τ to obtain the desired controllability for this problem.

In the above method, the measurement value of the air-fuel ratio sensor is used as it is for calculating the amount of fuel flowing into the cylinder. However, when the response delay of the sensor is large, the optimum X, 1 / τ is not estimated.

In this case, it is necessary to estimate X, 1 / τ in consideration of the response delay of the sensor. This can be dealt with by the following method.

First, a response delay model of the sensor is assumed to be a transfer function model of the fuel-air ratio. For example, assuming a first-order delay, the delay model is as follows.

The constant in equation (12) is determined such that the response of the sensor output when a predetermined input is applied to the sensor is measured, and the equation error of equation (12) is minimized based on the input / output time-series data. Equation (13) obtained as described above is converted to (3)
By making the expression and A / F of expression (10) equal to A / F _out , a relational expression between the fuel injection amount _Gf and the measured exhaust gas air-fuel ratio A / F _in can be obtained. The error equation of this relation is the parameter X,
Since it is linear with respect to 1 / τ, it is possible to estimate X, 1 / τ by the same method as described above.

This concludes the description of the operation of the parameter estimation means. Next, the operation of the parameter characteristic correcting means will be described. The parameter characteristic correcting means calculates the estimated parameter (k), Based on the various detected quantities X ₁ (k), X ₂ (k),... Of the engine which are the bases for the calculation of X, 1 / τ in the parameter calculating means, the predetermined various detected quantities X of the parameter calculating means are determined. _1, X _2, to modify the correspondence between ... and X, 1 / τ. As a correction method, for example, at the time k, in the state of X ₁ = S ₁ , X ₂ = S ₂ . X, 1 / τ calculated by the parameter calculating means is X ₀ , 1 / τ ₀ . At this time, the state X ₁ = S ₁ , X ₂ in the parameter calculation means
= S ₂ , ..., a new X, 1 / τ is calculated as X, 1 /
Let it be τ.

Here, X ₁ = S ₁ , X ₂ = S ₂ ... 0 <m <1 Here, x ₀ , 1 / τ ₀ is an estimated value at a time, The reason for not replacing is to prevent deterioration in controllability when estimation is not successful. If the parameter correction is repeated by the above method, the parameter gradually converges to the optimum value.

In this case, the estimated parameters are stored for each operation state for a predetermined time or a predetermined number, and when the predetermined time elapses or reaches the predetermined number, the average value of the estimated parameters is originally calculated. The correspondence between the various detection amounts and the control parameters (adhesion rate, evaporation rate) may be corrected by replacing the parameter values calculated by the parameter calculation means.

This concludes the description of the operation of the control system of the present invention. next,
The overall configuration of the control system and the operation of the control program when the above configuration is realized by a digital control unit will be described.

FIG. 2 shows an overall configuration diagram of the control system. Here, the L jetronics system is applied.

The control unit includes CUP, ROM, RAM, I / O, LSI, timer, and a bus for electrically connecting them. Signals from an air amount sensor, a water temperature sensor, a crank angle sensor, an air-fuel ratio sensor, and a throttle angle sensor are input to the I / O and LSI. Also, a signal to the fuel injection valve is output from the I / O, LSI. In addition, I / O, LS
I has an A / D converter on the input side and a D / A converter on the output side. The timer issues an interrupt request to the CPU at regular intervals, and in response to this request, the CPU executes a control program stored in the ROM.

Next, the operation of a control program for estimating parameters which are the characteristics of the present invention and correcting the correspondence between various detected amounts of the engine and X, 1 / τ based on the estimation results will be described with reference to FIG. .

First, in step 301, the rotational speed N at time k
From (k), k ₀ is calculated by equation (10).

Next, in step 302, the measured air amount Q _a (k) stored at the RAM at time k and the measured air-fuel ratio A / at time k + k ₀ are stored.
Read F (k + k ₀ ).

Next, in step 303, G
Calculate _fe (k) and store it in RAM.

Next, in step 304, G _f stored in the RAM
(I), G _fe (i) (i = k−2, k−1, k) (k), Is read, and (k + 1), And store it in RAM.

Next, in step 305,
The value of X, 1 / τ for various detected quantities x ₁ (k), x ₂ (k), x ₃ (k) of the engine, which are the bases for calculating X, 1 / τ at time k
Read X ₀ , 1 / τ ₀ .

Next, in step 306, the estimated value (k) of X, 1 / τ at time k stored in the RAM, Read (k).

Next, in step 307, x, 1 / τ
Is calculated, and the value is stored at the address where X ₀ , 1 / τ ₀ has been written.

 Next, in step 308, the value of k is increased by one.

Thus, the process is completed, and the process waits until the next interrupt request is issued.

In order to execute the above program, the amount of intake air
Q _a, the rotational speed N, the measured value of the air-fuel ratio A / F, the fuel injection amount G _f cylinder inflow fuel amount G _fe and, X, 1 / τ of the calculation of the underlying variable measurement values x _1, x _2, x The management of storage, deletion, etc. of the data of ₃ ,... Is required, but this is executed by another program.

Note that the above method does not adopt a configuration in which fuel control is performed by directly using the estimated value as shown in FIG. 4 because of concerns about the estimation time and estimation accuracy of the parameter X, 1 / τ. . Here, the problem of estimation time is that when the estimation delay is large, when the parameter X, 1 / τ is estimated, the parameter has already deviated from the optimal value in the operating state at that time, and sufficient control is performed. I can not get the nature. The problem of the estimation accuracy is that if a value that deviates greatly from the optimum value is estimated, the controllability is greatly deteriorated. If there are no above problems, the configuration shown in Fig. 4
M capacity can be greatly reduced, and the possibility of system downtime arises.

Next, an embodiment for the third means of the present invention will be described with reference to FIG.
This will be described with reference to FIG.

FIG. 5 shows transient characteristics of various air amounts at the time of acceleration and at the time of deceleration when targeting the L-Dietronics system. At the time of acceleration, the measured air amount, which is the basis for determining the fuel injection amount, is larger than the cylinder inflow air amount.
The reverse is true during deceleration. The control system to which the present invention is directed is intended to obtain the desired controllability by compensating only for the fuel delay, and does not take any consideration regarding the air amount. Therefore, in order to obtain the desired controllability, the air amount on which the fuel injection amount is determined must be the air amount flowing into the cylinder. However, actually, a measured air amount different from the cylinder inflow air amount is used. When determining the fuel injection amount based on the measured air amount, it is possible to obtain the desired controllability in a limited area by performing parameter matching such as adhesion rate and evaporation rate, but in various areas, Have difficulty. This is caused by the difference between the characteristics of the air amount at the time of acceleration and the characteristics at the time of deceleration in FIG. 5. In order to obtain the desired controllability (to set the air-fuel ratio to a target value) at the time of acceleration, parameter matching is performed. In other words, if the measured air amount is larger than the cylinder inflow air amount and the control based on the measured air amount causes the fuel supply amount to be larger than the required value, if the parameter matching is performed so that the fuel becomes slightly smaller, the control is performed at the time of deceleration. The air-fuel ratio, which is an amount, exceeds the target value. This is because when decelerating, the amount of air flowing into the cylinder is larger than the measured air amount, and in the control based on the measured air amount, since the fuel amount is smaller than the required value, the amount of fuel should be set slightly so that the parameters should be set. Because it is set to be less. That is, in the conventional control system, it is difficult to obtain desired controllability in both the operation modes during acceleration and deceleration. The above problems are the same in the D-jetronics system.

To solve this problem, in the present invention, for example, one of the output of each of the acceleration parameter calculation means, the deceleration parameter calculation means, and the output of each means is selected as the parameter calculation means as in the configuration of the control system shown in FIG. Signal selection means for sending the signal to the fuel injection control means is provided.

The acceleration parameter calculation means is a means for calculating a control parameter to obtain a desired controllability during deceleration, and the deceleration parameter calculation means is a means for calculating a control parameter to obtain a desired controllability during deceleration. is there. The correspondence between the above-mentioned respective means and various detection amounts of the operating state of the engine and the control parameters is determined in advance so that desired controllability is obtained. The signal selection means determines whether the vehicle is in an acceleration state or a deceleration state based on a deviation between the latest value of the throttle opening measured value and a value several hours before. If the vehicle is in the acceleration state, the acceleration parameter is calculated. If the output of the means is in a decelerating state,
The output of the deceleration parameter calculation means is sent to the fuel injection control means.

In the above configuration, separate parameters can be set at the time of acceleration and at the time of deceleration, so that extreme controllability can be obtained in various operation regions.

In the above embodiment, only two types of correspondence are provided, but a plurality of types may be provided depending on the magnitude of acceleration / deceleration.

〔The invention's effect〕

As described above, according to the present invention, if the number of operations increases, the values of the control parameters such as the adhesion rate and the deposition rate are automatically optimized, so that the number of steps for initial setting of the parameters can be reduced, and
There is an effect that adaptation to engine characteristics over time and individual differences becomes possible.

Further, by providing a plurality of relational expressions between the control parameters and the various detection amounts, there is an effect that desired controllability can be obtained in various operation regions.

[Brief description of the drawings]

FIG. 1 is a block diagram of the control system of the present invention, FIG. 2 is an overall block diagram of the control system when the present invention is realized by a digital control unit, and FIG. 3 is a flow of a control program which is a feature of the present invention. FIG. 4 is a block diagram of a control system in another embodiment of the present invention, FIG. 5 is a diagram showing transient characteristics of various air amounts, and FIG. 6 is a control system in still another embodiment of the present invention. FIG.

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Seiji Funabashi 1099 Ozenji Temple, Aso-ku, Kawasaki-shi, Kanagawa Prefecture Hitachi, Ltd. System Development Laboratory Co., Ltd. (56) References JP-A-60-36748 (JP, A) 63-246428 (JP, A) JP-A-60-162029 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) F02D 41/00 F02D 45/00

Claims (2)

(57) [Claims] 1. A fuel injection parameter is calculated from a detected value of an engine operation state based on a predetermined correspondence between an engine operation state and a parameter of a fuel flow in an intake pipe. In the adaptive control method for an engine, the exhaust air-fuel ratio, intake air amount, and rotation speed of the engine are detected, and the detected values of the detected exhaust air-fuel ratio, intake air amount, and rotation speed, and the time series data of the fuel injection amount are detected. From the adhesion rate,
Estimating the evaporation rate, calculating a new parameter using the parameter and the estimated value in the predetermined correspondence, correcting the correspondence using the new parameter, and setting the engine based on the corrected correspondence. An adaptive control method for an engine, comprising: controlling a fuel injection amount for a vehicle. 2. An adaptive control method for an engine, wherein the engine operation state in a predetermined correspondence relationship between the operation state of the engine and a flow in an intake pipe is at least one of a parameter indicating a state of an acceleration and a state of an air amount. 2. The adaptive control method for an engine according to claim 1, comprising: