JP2004263680A - Method and device for estimating engine misfire region, and method and device for adapting engine control parameter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To estimate a misfire region in an adapting process of an engine control parameter. <P>SOLUTION: A relation between the engine control parameter and an ignition delay/combustion duration and a relation between the ignition delay/combustion duration and combustion stability are modeled based on measurement data of an engine characteristic to provide a misfire estimating model, and a misfire region is estimated using the misfire estimating model. When the misfire region can be thus estimated using the misfire estimating model, measuring points are arranged only in a combustible region except for the estimated misfire region, the engine characteristic is measured, and an engine characteristic model is prepared based on the measurement data. When the engine characteristic model satisfies a required accuracy, the control parameter is adapted using the engine characteristic model. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an engine misfire area estimation method and an engine control parameter adaptation method for estimating an engine misfire area in an engine control parameter adaptation process, and an engine misfire area estimation apparatus and an engine control parameter adaptation apparatus.
[0002]
[Prior art]
In recent years, engines installed in automobiles have realized improvements in output, reduction of exhaust emissions, reduction of fuel consumption, etc. by controlling control parameters such as ignition timing and injection timing to optimal values using maps in accordance with operating conditions using an on-board computer. I am trying to do it. Since the optimal value of the map parameter of the control parameter of the engine differs for each model, it is necessary to adapt the map parameter of the control parameter to satisfy the required performance in the process of designing and developing the engine.
[0003]
In the conventional adaptation method, for example, when the two control parameters of the ignition timing and the injection timing are adapted, the required number of measurement points are set for each of the ignition timing and the injection timing, and the measurement conditions of all combinations thereof are set. (If the number of measurement points of each control parameter is 10, for example, 10 × 10 = 100 conditions), the engine is operated, torque, exhaust emission, fuel efficiency, etc. are measured to evaluate the degree of conformity. It seeks the optimal point with the highest fitness.
[0004]
However, since recent high-performance engines are equipped with various functions such as a variable valve timing mechanism and an EGR system, the control parameters to be applied are not only the ignition timing and the injection timing, but also the valve timing and the EGR rate. Need to fit. For this reason, the number of control parameters to be adapted tends to increase, and the work of adjusting the control parameters has become extremely troublesome.
[0005]
Therefore, as shown in Patent Document 1 (Japanese Patent Application Laid-Open No. 2002-206456), the characteristic values of the engine are measured at a predetermined number of representative measurement points set in advance, and each control parameter and the engine value are determined based on the measurement result. It has been proposed to obtain a model formula that defines a relationship with a characteristic value and calculate an appropriate value of a control parameter using the model formula.
[0006]
[Patent Document 1]
JP-A-2002-206456 (Pages 1 and 2 etc.)
[0007]
[Problems to be solved by the invention]
However, in an engine equipped with various functions such as a variable valve timing mechanism and an EGR system, many control parameters are intricately entangled to change the combustion state. It is difficult to model accurately. For this reason, the adaptation value of the control parameter is calculated using a model equation with low accuracy, and the calculation accuracy of the adaptation value of the control parameter is low.
[0008]
As a countermeasure against this, in Patent Document 1 described above, an appropriate value of a control parameter calculated by a model equation is evaluated in an actual vehicle running test, the model equation is corrected, and an appropriate value of the control parameter is calculated again to execute an actual vehicle running test. By repeating the operation of evaluating the control parameter many times, the accuracy of the appropriate value of the control parameter is ensured.
[0009]
However, in this method, it is necessary to repeat the actual vehicle running test many times, and as a result, the effect of reducing the number of adaptation steps (adaptation time) by introducing the model formula may be considerably reduced.
[0010]
SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and therefore has as its object to provide a method for estimating an engine misfire area and a method for adapting an engine control parameter which can be expected to greatly reduce the time required for the operation of adjusting the control parameters of the engine. Another object of the present invention is to provide an engine misfire area estimating apparatus and an engine control parameter adapting apparatus.
[0011]
[Means for Solving the Problems]
In order to achieve the above object, the present invention models a relationship between an engine control parameter and an ignition delay / combustion period and a relationship between an ignition delay / combustion period and combustion stability based on measurement data of engine characteristics. Thus, a misfire estimation model is created, and a misfire area is estimated using the misfire estimation model (claims 1 and 9).
[0012]
As described above, if the misfire region can be estimated from the misfire estimation model, measurement points are arranged in the combustible region excluding the estimated misfire region to measure engine characteristics, and an engine characteristic model is created based on the measurement data. The control parameters can be adapted using this engine characteristic model (claims 8 and 11). This makes it possible to remove the measurement points included in the misfiring region in advance and to measure the engine characteristics only in the combustible region in the process of adjusting the control parameters, thereby greatly reducing the measurement time, The operation time of the parameter matching process can be greatly reduced.
[0013]
The misfire estimation model used in the present invention, when modeling the relationship between control parameters and combustion stability (degree of misfire), estimates the ignition delay and the combustion period from the control parameters as shown in FIG. ▼, a model (2) for estimating the combustion stability (degree of misfire) from the ignition delay and the combustion period was created, and a misfire estimation model was constructed from these two models (1) and (2). .
[0014]
In general, the normal combustion ends when spark discharge of the spark plug grows in the flame nucleus and the flame propagates to the entire air-fuel mixture. Analyzing the mechanism of misfire, on the other hand, misfires can occur during the ignition delay during which the flame is extinguished without spark discharge growing to the flame nucleus and during the combustion period when the flame quenches after the flame nucleus occurs. There are two types of misfires (see FIGS. 2 and 3). Focusing on these two types of misfire occurrence mechanisms, the misfire estimation model of the present invention first uses two parameters, “ignition delay” and “combustion period”, as parameters for evaluating combustion stability (degree of misfire). The combustion stability (degree of misfire) is estimated from the estimated "ignition delay" and "combustion period". With this configuration, it is possible to accurately estimate both types of misfires, that is, misfire during the ignition delay and misfire during the combustion period, and improve the estimation accuracy of the misfire region.
[0015]
Here, the ignition delay and the combustion period are defined by the combustion mass ratio obtained from the combustion pressure waveform shown in FIG. 2, and a period corresponding to, for example, 0 to 10% of the combustion mass ratio is called an ignition delay, and the combustion mass ratio For example, a period corresponding to 10 to 90% is called a combustion period (see FIG. 3).
[0016]
In order to improve the accuracy of the misfire estimation model, the generation of the misfire estimation model may be repeated an appropriate number of times. Specifically, after estimating the misfire area using the created misfire estimation model, the required number of measurement points are arranged in the combustible area excluding the estimated misfire area, and the engine characteristics are measured. The accuracy of the misfire estimation model may be improved by repeating the process of creating the misfire estimation model using. By doing so, the accuracy of the misfire estimation model can be improved each time the misfire estimation model is repeatedly created.
[0017]
In this case, when arranging measurement points for preparing a misfire estimation model, a required number of measurement points may be arranged in the combustible region by an experiment design method (claim 3). After arranging a predetermined number of measurement points by the experimental design within the variable range of the control parameter, a measurement point estimated to be within the misfire area by the misfire estimation model may be taken from among the measurement points ( Claim 4). At this time, the process of adding new measurement points for the number of measurement points estimated and removed within the misfire area by the experimental design method is repeated until the required number of measurement points are arranged in the combustible area. (Claim 5). In short, it is only necessary that the required number of measurement points can be finally arranged in the combustible region.
[0018]
When estimating the ignition delay using the misfire estimation model, the ignition delay may be estimated based on at least the ignition timing and the composition of the in-cylinder gas. Among the various control parameters of the engine, the parameter that most affects the ignition delay is the ignition timing and the composition of the in-cylinder gas, so that the ignition delay is estimated based on at least the ignition timing and the composition of the in-cylinder gas. Then, the ignition delay can be accurately estimated.
[0019]
Further, when estimating the combustion period using the misfire estimation model, the combustion period may be estimated based on at least the combustion timing and the composition of the in-cylinder gas. Among the various control parameters of the engine, the parameter that most affects the combustion period is the combustion timing. Therefore, if the combustion period is estimated based on at least the combustion timing, the combustion period can be accurately estimated. be able to.
[0020]
BEST MODE FOR CARRYING OUT THE INVENTION
<< Embodiment (1) >>
Hereinafter, an embodiment (1) of the present invention will be described with reference to FIGS.
First, an outline of the misfire estimation model of the embodiment (1) will be described with reference to FIG. The misfire estimation model of this embodiment (1) is composed of two models (1) and (2). The first model (1) models the relationship between the control parameters of the engine and the ignition delay / combustion period. The second model (2) models the relationship between ignition delay / burning period and combustion stability.
[0021]
Here, the ignition delay and the combustion period are defined by the combustion mass ratio obtained from the combustion pressure waveform shown in FIG. 2, and as shown in FIG. The period corresponding to, for example, 10 to 90% of the combustion mass ratio is referred to as a combustion period. The combustion mass ratio is the ratio of the mass of the burned gas to the total mass of the in-cylinder gas immediately before the start of combustion.By setting the horizontal axis of the combustion mass ratio graph to the crank angle, the ignition delay and the combustion period are reduced. , Each represented by a crank angle.
[0022]
The combustion stability indicates the degree of misfire, and the better the combustion stability, the smaller the variation in combustion and the more difficult the misfire. In the present embodiment, for example, COV (Coefficiency Of Variation) is used as a parameter indicating combustion stability. This COV is obtained by dividing the standard deviation of the indicated mean effective pressure by the average value.
[0023]
Next, a specific configuration of the misfire estimation model will be described based on FIG. As described above, the misfire estimation model consists of two models (1) and (2). The first model (1) models the relationship between the control parameters of the engine and the ignition delay and combustion period. The second model {circle around (2)} models the relationship between ignition delay / burning period and combustion stability. Further, the first model (1) includes a G / F calculation unit 11 for calculating the in-cylinder gas air-fuel ratio G / F, an ignition delay estimation unit 12 for estimating the ignition delay, and a combustion mass ratio of 50% (MBF50). %) And an MBF 50% estimating unit 13 for estimating the crank angle and a combustion period estimating unit 14 for estimating the combustion period.
[0024]
The G / F calculation unit 11 calculates the in-cylinder gas air-fuel ratio G / F using a predetermined function or map based on the intake air amount, supply air-fuel ratio (fuel injection amount), EGR rate, and the like, which are control parameters of the engine. I do. The in-cylinder gas air-fuel ratio G / F is used as a parameter relating to the composition of the in-cylinder gas.
[0025]
Generally, the ignition delay and the combustion period tend to be longer as the in-cylinder gas air-fuel ratio G / F becomes leaner. Further, the ignition delay tends to become shorter as the ignition timing approaches TDC (compression top dead center). Further, the combustion period tends to be shorter as TDC approaches the combustion timing. Therefore, the ignition delay estimating unit 12 calculates the ignition delay based on the in-cylinder gas air-fuel ratio G / F and the ignition timing by using a predetermined function, a map, or the like.
[0026]
In addition, even with the same ignition timing, the combustion timing changes depending on the length of the ignition delay, and the effect on the combustion period is different. Therefore, instead of the ignition timing, the crank angle at the reaction, which is the combustion timing, is changed. It seems appropriate to use. Therefore, the combustion period estimating unit 14 uses a crank angle of 50% (MBF50%), which is the center of the combustion mass ratio of 10 to 90%, as the crank angle at the time of the reaction. The combustion period is calculated by a predetermined function or map based on the in-cylinder gas air-fuel ratio G / F.
[0027]
The MBF 50% estimating unit 13 calculates the crank angle of the MBF 50% based on the crank angle of the combustion mass ratio 10% (MBF 10%) and the in-cylinder gas air-fuel ratio G / F using a predetermined function or map. Here, the crank angle of 10% of the MBF is obtained by adding the ignition delay to the ignition timing.
[0028]
The second model (2) calculates COV as a parameter representing combustion stability by using a predetermined function or map based on the ignition delay and the combustion period estimated by the first model (1). Ultimately, it is determined whether or not this is a misfire region based on whether or not the value of COV is equal to or more than a predetermined value (for example, 10% or more).
[0029]
A system configuration of an engine misfire area estimating apparatus (engine control parameter adapting apparatus) for estimating an engine misfire area using the misfire estimation model described above will be described with reference to FIG.
[0030]
A suitable engine 21 is mounted on a bench 22 and the crankshaft of this engine 21 is connected to a dynamometer 23. During the misfire area estimation and adaptation operation, various actuators of the engine 21 are controlled by an electronic control unit (ECU) 24. The electronic control unit 24 is connected to a computer 27 via an adaptation tool 26 (adaptation means), and sends a control signal from the computer 27 to the electronic control unit 24 via the adaptation tool 26 during misfire area estimation / adaptation work. By transmitting, the map constant of each control parameter in the electronic control unit 24 is changed. The throttle opening of the engine 21 during the misfire area estimation and adaptation work is adjusted by the throttle control device 25.
[0031]
During the misfire area estimation / adaptation work, the dynamometer control panel controls the dynamometer 23 and the throttle control device 25 to control the engine load, and transmits the engine torque measured by the dynamometer 23 to the computer 27. Each cylinder of the engine 21 is provided with an in-cylinder pressure sensor (not shown) for detecting an in-cylinder pressure (combustion pressure). An output signal of the in-cylinder pressure sensor of each cylinder is analyzed by a combustion analyzer 29, and a combustion pressure waveform is obtained. Is measured to calculate a combustion mass ratio (MBF) and the like. The exhaust gas discharged from the engine 21 is analyzed by the exhaust gas analyzer 30, and the measurement result of the emission of NOx, CO, HC, etc. in the exhaust gas is transmitted to the computer 27. The engine 21 applicable to this system may be any system such as an intake port injection engine and a direct injection engine.
[0032]
Using the engine misfire area estimating apparatus (engine control parameter adapting apparatus) configured as described above, a misfire estimation model is created as follows by the misfire estimation model creation program shown in FIG. This program is executed by the computer 27, and plays a role as a model creating means referred to in the claims.
[0033]
When the program is started, first, in step 101, the minimum measurement points necessary for creating a misfire estimation model are arranged by an experiment design method or the like, and the engine 21 is operated under the conditions of each measurement point. Various engine characteristic data such as an intake air amount, a supply air-fuel ratio, an EGR rate, an ignition timing, and a combustion pressure waveform are measured. At this stage, since the misfire area cannot be estimated yet, the measurement points may be arranged within the variable range of the engine control parameters by an experiment design method or the like. If there is a misfire estimation model created by a similar engine, the necessary number of measurement points may be arranged in the combustible region estimated using the misfire estimation model by an experiment design method or the like.
[0034]
In the next step 102, a misfire estimation model is created based on the measurement data. At this time, the in-cylinder gas air-fuel ratio G / F is calculated from the intake air amount, the supplied air-fuel ratio, the EGR rate, etc., the combustion mass ratio (MBF) is calculated from the combustion pressure waveform, and the ignition delay (0 to 10 of MBF) is calculated. %), The combustion period (10-90% of MBF), the crank angle of MBF 10%, the crank angle of MBF 50%, COV, etc. are calculated, and variables of the misfire estimation model are identified from these data. Thereby, a misfire estimation model is created.
[0035]
At the stage of arranging the measurement points in step 101, since the misfire area cannot be estimated yet, it is considered that some measurement points are included in the misfire area. Therefore, it seems that the accuracy of the engine characteristic model created excluding the misfire point does not satisfy the required level.
[0036]
Therefore, when the accuracy of the engine characteristic model does not satisfy the required accuracy, the measurement is repeated until the required accuracy is satisfied (step 108). At this time, the misfire estimation model improves accuracy by using data measured for the previous engine characteristic.
[0037]
After the misfire estimation model is created, the combustible area (misfire area) is estimated by the previous misfire estimation model (step 104), and the required number of measurement points are arranged in the estimated combustible area by an experiment design method or the like (step 104). 105). Thereafter, the engine 21 is operated under the conditions of each measurement point, and various engine characteristic data such as an intake air amount, a supply air-fuel ratio, an EGR rate, an ignition timing, and a combustion pressure waveform are measured (step 106). An engine characteristic model is created based on the measurement data (step 107).
[0038]
Thereafter, the accuracy of the engine characteristic model is verified (step 108). If the required accuracy is not satisfied, the process returns to step 102 and the process of creating a misfire estimation model based on the measured data of the engine characteristics is repeated. At this stage, the combustible region (misfire region) is estimated using the previously created misfire estimation model, and the measurement points can be arranged only in the combustible region. Therefore, the accuracy of the misfire estimation model can be improved.
[0039]
By doing so, the accuracy of the misfire estimation model can be improved each time the misfire estimation model is repeatedly created. Then, when the required accuracy of the engine characteristic model is satisfied, the creation of the misfire estimation model ends.
[0040]
Thereafter, when the control parameters of the engine 21 are adapted, a combustible region (misfire region) is estimated using the misfire estimation model, and measurement points are arranged only within the combustible region to measure engine characteristics. An engine characteristic model is created based on the measured data, and the control parameters are adapted using the engine characteristic model (step 109).
[0041]
This makes it possible to remove the measurement points included in the misfiring region in advance and to measure the engine characteristics only in the combustible region in the process of adjusting the control parameters, thereby greatly reducing the measurement time, The operation time of the parameter matching process can be greatly reduced.
[0042]
<< Embodiment (2) >>
In the above embodiment (1), when the generation of the misfire estimation model is repeated, the combustible area (misfire area) is estimated using the previously generated misfire estimation model, and the required number of measurement points is set only within the combustible area. However, after arranging a predetermined number of measurement points within the variable range of the control parameter by the experimental design method, the measurement estimated from the measurement points to be within the misfire area by the previous misfire estimation model Points may be taken. At this time, the process of adding new measurement points for the number of measurement points estimated and removed within the misfire area by the experimental design method is repeated until the required number of measurement points are arranged in the combustible area. It is good to
Hereinafter, a program for creating a misfire estimation model according to the embodiment (2) of the present invention will be described with reference to FIG.
[0043]
In the misfire estimation model creation program of this embodiment (2), the necessary number of measurement points are arranged in the estimated combustible region by the experiment design method or the like in the same manner as in the embodiment (1) (step 201). 204), a misfire area is estimated by a previous misfire estimation model, and measurement points existing in the misfire area are deleted (step 205). Thereafter, the process proceeds to step 206, where it is determined whether or not the number of remaining measurement points (measurement points in the combustible region) that have not been deleted is equal to or greater than the required number. New measurement points corresponding to the number of measurement points estimated and removed in the misfire area in the above step 205 are added by the experiment design method, and the process returns to step 205, and the added measurement points are present in the misfire area. Delete a measurement point.
[0044]
The above processing is repeated until the number of measurement points in the combustible region becomes equal to or more than the necessary number. When the number of measurement points becomes equal to or more than the required number, the process proceeds to step 208, and the engine 21 is operated under the conditions of each measurement point. Various engine characteristic data such as an intake air amount, a supply air-fuel ratio, an EGR rate, an ignition timing, and a combustion pressure waveform are measured, and an engine characteristic model is created based on the measured data (step 209).
[0045]
Thereafter, the accuracy of the engine characteristic model is verified (step 210). If the required accuracy is not satisfied, the process returns to step 202 and the process of creating a misfire estimation model based on the measured data of the engine characteristics is repeated. Thereafter, the control parameters are adapted using the engine characteristic model in the same manner as in the embodiment (1) (step 211).
[0046]
In the embodiment (2) described above, the same effect as in the embodiment (1) can be obtained.
[Brief description of the drawings]
FIG. 1 is a diagram schematically showing a misfire estimation model of the present invention; FIG. 2 is a diagram showing a combustion pressure waveform; FIG. 3 is a diagram showing a relationship between a combustion mass ratio and a crank angle; FIG. FIG. 5 is a diagram schematically showing a misfire estimation model of mode (1). FIG. 5 is a diagram showing a system configuration of an engine misfire region estimation device. FIG. 6 is a diagram showing processing of a misfire estimation model creation program according to an embodiment (1) of the present invention. FIG. 7 is a flowchart showing the flow of processing of the misfire estimation model creation program according to the embodiment (2) of the present invention.
11: G / F calculation unit, 12: ignition delay estimation unit, 13: MBF 50% estimation unit, 14: combustion period estimation unit, 21: engine, 26: adaptation tool (adaptation unit), 27: computer (model creation unit, Misfire area estimation means).

Claims (11)

A method for estimating a misfire area of an engine in a process of adjusting control parameters of an engine,
Based on the measured data of the engine characteristics, model the relationship between the control parameters of the engine and the ignition delay / combustion period and the relationship between the ignition delay / combustion period and the combustion stability to create a misfire estimation model. A method for estimating an engine misfire area, comprising estimating a misfire area using a computer. After estimating the misfire area using the misfire estimation model, a required number of measurement points are arranged in a combustible area excluding the estimated misfire area to measure engine characteristics, and the misfire estimation is performed using the measurement data. The engine misfire region estimation method according to claim 1, wherein the accuracy of the misfire estimation model is improved by repeating a process of creating a model. 3. The engine misfire area estimation method according to claim 2, wherein when arranging the measurement points, a required number of measurement points are arranged in the combustible area by an experiment design method. When arranging the measurement points, after arranging a predetermined number of measurement points by an experiment design method within the variable range of the control parameter, it is estimated that the misfire estimation model is within the misfire area from among those measurement points. 3. The method for estimating an engine misfire area according to claim 2, wherein measurement points are taken. The process of adding new measurement points for the number of measurement points estimated and removed in the misfire area by the experimental design method is repeated until the required number of measurement points are arranged in the combustible area. 5. The method for estimating an engine misfire area according to claim 4, wherein: The engine misfire region estimating method according to any one of claims 1 to 5, wherein the misfire estimation model is configured to estimate an ignition delay based on at least an ignition timing and a composition of in-cylinder gas. . The engine misfire region estimation according to any one of claims 1 to 6, wherein the misfire estimation model is configured to estimate a combustion period based on at least a combustion timing and a composition of in-cylinder gas. Method. After estimating a misfire area using the misfire area estimation method according to any one of claims 1 to 7, a measurement point is arranged in a combustible area excluding the estimated misfire area to measure engine characteristics. An engine control parameter matching method, wherein an engine characteristic model is created based on measurement data, and the control parameter is adapted using the engine characteristic model. An engine misfire area estimating apparatus for estimating a misfire area of an engine,
Model creating means for creating a misfire estimation model by modeling the relationship between the control parameters of the engine and the ignition delay / combustion period based on the measured data of the engine characteristics, and the relationship between the ignition delay / combustion period and the combustion stability,
An engine misfire area estimating device, comprising: a misfire area estimating means for estimating a misfire area using the misfire estimation model. The model creating means estimates a misfire area using the created misfire estimation model, and then arranges a required number of measurement points in a combustible area excluding the estimated misfire area to measure engine characteristics. The engine misfire region estimating apparatus according to claim 9, wherein the accuracy of the misfire estimation model is improved by repeating a process of creating the misfire estimation model based on the measurement data. An engine misfire area estimating device according to claim 9 or 10,
The engine characteristics are measured by arranging measurement points in the combustible region excluding the misfire region estimated by the engine misfire region estimation device, and an engine characteristic model is created based on the measured data. And an adapting means for adapting the control parameter.

Images