JP2006017698A - Setting method of experimental planning for engine control parameter, program for making computer execute the setting method of experimental planning, computer-readable recording medium storing the program, and setting apparatus of experimental planning for engine control parameter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a setting method of experimental planning for engine control parameters, which enables the engine control parameters to be adapted in a short period of time. <P>SOLUTION: The setting method of experimental planning for the engine control parameters comprises a step (S1) of setting a first prescribed number of experimental points that are required in order to obtain a response curved surface, with each being expressed by a combination of a plurality of engine control parameters; a step (S2) of inputting the experiment result data measured at the first prescribed number of experimental points being set; a step (S3) of extracting effective experimental points, from which effective data for the response curved surface are obtained, and missing values from which effective data are not obtained, while referring to the experiment result data; and steps (S5, S6) of setting a second prescribed number of additional experimental points which are used, in combination with the effective experimental points after eliminating the missing values in the experimental points, in the case the number of the missing values exceeds a prescribed threshold level, thereby enabling the response curved surface to be obtained. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an engine control parameter experiment plan setting method, a program for causing a computer to execute the experiment plan setting method, a computer-readable recording medium storing the program, and an engine control parameter experiment plan setting apparatus.

  In an automobile engine, complicated control is performed so that engine characteristics such as exhaust gas characteristics, fuel consumption characteristics, stable combustibility, and power performance satisfy various requirements.

  In other words, electronic control for engine control is set in advance by setting appropriate values for each control parameter such as the optimum fuel injection amount and optimum ignition timing according to the engine operating state determined based on the engine speed and load. Stored in the unit (ECU). Then, the ECU controls the engine while referring to the set adaptation value. Different requirements for each engine characteristic are met by using adapted values corresponding respectively to different operating conditions.

  FIG. 22 is a diagram for generally explaining the adaptation values of the engine control parameters.

  FIG. 22 shows how each engine characteristic changes according to the ignition timing of the engine under a condition where a predetermined engine rotation speed and a predetermined load are applied. In FIG. 22, NOx represents the concentration of nitrogen oxides, which is one of the exhaust gas characteristics, TF represents torque fluctuation representing the characteristics of power performance, and BSFC represents fuel consumption, which is a fuel consumption characteristic.

Focusing only on the fuel consumption BSFC, the point with the best fuel efficiency is the point C in FIG. However, since the engine characteristic requires that the torque fluctuation TF is TF ₀ or less, the ignition timing is limited to the range of A. Further, since there is a demand that the nitrogen oxide concentration NOx in the exhaust gas must be NOx ₀ or less, the ignition timing is limited to the range of B. A point D that is in a range that satisfies A and B and that has the smallest fuel consumption BSFC is an adaptation point of ignition timing that is a required control parameter.

  As described above, in order to find a conforming point that satisfies the conditions of clean exhaust gas, good fuel consumption, stable combustion, and good power performance, perform experiments efficiently while changing each of the engine control parameters. Is required.

  In practice, it is difficult to continuously capture the corresponding data by continuously changing the control parameters (in the case of FIG. 22, the ignition timing). Therefore, by performing experiments at a plurality of experimental points and measuring data, and modeling with an approximate expression based on the data, engine characteristics as shown by the three curves in FIG. 22 are obtained.

  Then, by analyzing this engine characteristic, a matching point (point D in the case of FIG. 22) corresponding to the position of the accelerator pedal operated by the driver when a certain engine load is applied at a certain vehicle speed is obtained. An engine that satisfies various engine characteristics can be realized by storing a control map, which is a set of matching points at various vehicle speeds and loads, in an ECU that performs engine control.

  However, in recent years, the number of engine control parameters that must be taken into account has increased, and the conventional method based on trial and error requires an enormous number of experiments and associated time. For example, in recent years, with the improvement in engine performance, a valve timing variable mechanism that changes the valve timing of an intake valve or the like as an engine parameter has been adopted, and exhaust gas is reduced in order to reduce the amount of nitrogen oxides generated in the exhaust gas. EGR is used to return a part of the engine to the intake system, and an in-cylinder injection engine has been put into practical use.

  As described above, in addition to the ignition timing and the fuel injection amount, various engine control parameters such as the variable valve timing, the ratio of mixing the exhaust with the intake air, the fuel injection timing, and the fuel pressure must be taken into consideration.

On the other hand, different standards may be applied to exhaust gas regulations in each country, and it is necessary to adjust each concentration of NOx, CO ₂ , HC, etc. in the exhaust gas so as to satisfy the regulation values of each country. In other words, the required engine characteristics are also complicated.

  By simply overlapping the points at which the performance of each control parameter is the best, an appropriate value cannot be obtained. This is because there is an interaction between one engine control parameter and another engine control parameter, which influence each other.

  In the case of a conventional engine having only three engine control parameters, it was possible to find a matching point by repeating trial and error. However, as the performance of the engine has increased, the number of engine control parameters has increased to five. If it is increased to 7 or so, it has become almost impossible for human beings to find a matching point by trial and error in the traditional sense.

Therefore, in recent years, in order to increase the efficiency of adaptation of engine control parameters, as shown in, for example, Japanese Patent Application Laid-Open No. 2002-322938 (Patent Document 1), the experimental design method has been used for automobile experiments. ing.
JP 2002-322938 A

  Various optimization methods include the Simplex method, Newton method, and recently optimization methods based on genetic algorithms. These experiments are carried out within the measurable range, and experimental data necessary for optimization of control parameters is obtained. To get. However, because there are many experimental points, the efficiency is low, and a demand for a new method with higher efficiency is increasing.

  Thus, as described above, the use of the experimental design method for automobile experiments is also being studied. However, when the experimental design method is used, points that cannot be measured among the planned experimental points (hereinafter referred to as this). Is called a missing point), the prediction accuracy is greatly deteriorated. Therefore, in order to obtain an accurate model of engine characteristics, if there are several missing points, it is necessary to re-design the experiment plan so that normal data can be obtained at each experimental point.

  In such a case, if an area that cannot be measured, such as an engine misfire area or an overheated area, can be predicted in advance, an experimental plan can be made with the area removed from the beginning. However, in reality, an area that cannot be measured is often found by performing an experiment. For this reason, if there are a large number of experimental points belonging to a region that cannot be measured, that is, missing points, among the set experimental points, it is necessary to reestablish and remeasure the experiment design method because the measured data is incomplete. In such a case, it becomes a factor that greatly delays the stage of adapting the engine control parameters in engine development.

  An object of the present invention is to provide an engine control parameter experiment plan setting method capable of adapting engine control parameters in a short period of time, a program for causing a computer to execute the experiment plan setting method, and a computer readable recording of the program It is to provide an experiment plan setting device for a simple recording medium and engine control parameters.

  In summary, the present invention is an engine control parameter experimental plan setting method, in which experimental points each indicated by a combination of a plurality of engine control parameters are set by a first predetermined number necessary to obtain a response surface. A step of inputting the experimental result data measured at the set first predetermined number of experimental points, and an effective data obtained for obtaining a response surface with reference to the experimental result data. The step of extracting missing points for which experimental data and valid data could not be obtained, and when the number of missing points exceeds a predetermined threshold, Setting a second predetermined number of additional experimental points that can be used in combination with effective experimental points to obtain a response surface.

Preferably, the step of setting the additional experimental points includes a matrix created using a plurality of engine control parameters corresponding to each effective experimental point and a plurality of engine control parameters corresponding to the second predetermined number of additional experimental points. When X, the step of obtaining a combination of additional experimental points that maximizes the determinant | X ^T X |

  Preferably, the first predetermined number of experimental points is determined based on the central composite design.

  More preferably, the experimental candidate points when setting the first predetermined number of experimental points are the maximum value, minimum value, and intermediate value between the maximum value and the minimum value of the control range of the engine control parameter. Any one of them is determined by the combination of selecting each engine control parameter, and the additional experimental points are the three selected values including the point between the maximum value and the intermediate value and the point between the minimum value and the intermediate value. It is selected from the combinations in which any one value is selected for each engine control parameter from the added five selected values, and the missing points are removed.

  Preferably, the engine control parameter experimental design setting method is such that, in the step of setting only the first predetermined number necessary for obtaining the response surface, if a certain engine control parameter can be expressed by two or more other parameters, the experimental point The step of setting as an experimental point only when the engine control parameter at is in the operating range of the other two or more parameters, and when the engine control parameter at the experimental point is outside the operating range of the other two or more parameters And a step of making a missing point.

  In another aspect, the present invention is a program that causes a computer to execute any one of the engine control parameter experiment plan setting methods described above.

  In still another aspect, the present invention provides a computer-readable recording medium on which a program for causing a computer to execute any one of the engine control parameter experiment plan setting methods is recorded.

  According to still another aspect of the present invention, there is provided an engine control parameter experiment plan setting apparatus, wherein a first predetermined value necessary for obtaining a response surface is obtained from experimental points each indicated by a combination of a plurality of engine control parameters. The first setting unit for setting the number, the input unit for inputting the experimental result data measured at the set first predetermined number of experimental points, and obtaining the response surface with reference to the experimental result data An extraction unit that extracts valid experimental points for which valid data was obtained and missing points for which valid data was not obtained, and missing points when the number of missing points exceeds a predetermined threshold And a second setting unit for setting a second predetermined number of additional experimental points that can obtain a response surface by using in combination with the effective experimental points.

Preferably, the second setting unit defines a matrix created using a plurality of engine control parameters corresponding to each effective experimental point and a plurality of engine control parameters corresponding to the second predetermined number of additional experimental points as X. Then, a combination of additional experimental points that maximizes the determinant | X ^T X |

  Preferably, the first predetermined number of experimental points is determined based on the central composite design.

  More preferably, the experimental candidate points when setting the first predetermined number of experimental points are the maximum value, minimum value, and intermediate value between the maximum value and the minimum value of the control range of the engine control parameter. Any one of them is determined by the combination of selecting each engine control parameter, and the additional experimental points are the three selected values including the point between the maximum value and the intermediate value and the point between the minimum value and the intermediate value. It is selected from the combinations in which any one value is selected for each engine control parameter from the added five selected values, and the missing points are removed.

  According to the present invention, when there is a missing point such as an automobile engine, the accuracy of the model of each characteristic is efficiently improved by performing a few additional experiments while making use of normal data already obtained through experiments. It can be secured. As a result, the time required for adapting the engine control parameters can be shortened.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and description thereof will not be repeated.

[Configuration of engine control parameter conforming system]
FIG. 1 is a block diagram showing a configuration of an engine control parameter adaptation system to which an engine control parameter experimental plan setting method according to an embodiment of the present invention is applied.

  The target controlled by the control parameter is the in-cylinder injection type gasoline engine 10. Referring to FIG. 1, a gasoline engine 10 includes a cylinder 11, a piston 12, an injector 14 that is disposed above a combustion chamber 13 defined by the cylinder 11 and the piston 12, and directly injects fuel, and a combustion chamber. A spark plug 15 for igniting the air-fuel mixture in 13 is provided.

  Air is sucked into the combustion chamber 13 from the intake passage 16, and fuel is injected into this to become an air-fuel mixture. Combustion gas in which the air-fuel mixture is combusted by ignition is discharged from the combustion chamber 13 to the exhaust passage 17 as exhaust. Each timing of intake of air from the intake passage 16 and discharge of exhaust to the exhaust passage 17 is set by opening timings of the intake valve 18 and the exhaust valve 19, respectively. In the case of the engine 10 shown in FIG. 1, the valve opening timing of the intake valve 18 (the amount of overlap with the valve opening period of the exhaust valve 19) is variably set by the variable valve timing mechanism 20.

  On the other hand, the amount of air taken into the combustion chamber 13 of the engine 10 is adjusted by an electronically controlled throttle 21 provided in the intake passage 16. A part of the exhaust discharged to the exhaust passage 17 is returned to the intake passage 16 via the EGR passage 22. The returned exhaust amount is adjusted by the valve opening amount of the EGR valve 23.

  Such control of the engine 10 is performed by the ECU 30. In addition, information from various sensors that measure the operating state of the engine, such as the water temperature sensor 26 and the rotation speed sensor 25 provided in the vicinity of the output shaft 24 of the engine 10, is input to the ECU 30 as measurement information.

  The adaptive system of the present embodiment calculates each adaptive value of the control map that sets various control parameters of the engine 10 to appropriate values. The adaptation system of the present embodiment includes a dynamometer 31 connected to the output shaft 24 of the engine 10, a dynamometer operation panel 32 for operating the dynamometer 31, and a dynamometer operation panel for controlling the dynamometer 31 to predetermined conditions. And an automatic measuring device 33 that sends a command to 32.

  Here, the dynamometer 31 absorbs torque generated by the output shaft 24 of the engine 10. As a result, various tests can be performed with the engine 10 mounted in a pseudo state on the vehicle. Then, the torque absorbed by the dynamometer 31 is controlled by operating the dynamo operation panel 32 in accordance with a command from the automatic measuring device 33.

  The adaptation system of the present embodiment further includes a panel checker 34 that mediates data exchange between the ECU 30 and the automatic measurement device 33. Then, the automatic measuring device 33 acquires measurement information of the engine 10 held in the ECU 30 via the panel checker 34.

  When the engine 10 is actually mounted on a vehicle, the driving state is controlled based on measurement information input to the ECU 30 from various sensors or the like. On the other hand, when the dynamometer 31 is used to create a state of being mounted on the vehicle in a pseudo manner, data such as the accelerator pedal depression amount reflecting the driver's will is not supplied to the ECU 30.

  Accordingly, the automatic measuring device 33 supplies the ECU 30 with data corresponding to the amount of depression of the accelerator pedal via the panel checker 34 while monitoring the state of the engine 10 with reference to the measurement information of the engine 10. In this way, the automatic measuring device 33 controls the engine 10 to a desired operating state.

  On the other hand, in the ECU 30, a control map that can roughly control the engine 10 such as a control map of an engine of a model similar to the engine 10 is stored as control information of the engine 10.

  A command for controlling the engine 10 and the dynamometer 31 by the automatic measuring device 33 is largely set based on a condition file in the automatic measuring device 33. In this condition file, basically, control parameters (ignition timing, fuel injection amount, variable valve timing, exhaust gas are mixed with intake air) for each operating state (rotation speed and load) of the engine 10 desired to be measured. , The fuel injection timing, fuel pressure, etc.) are written. The engine 10 is fixedly controlled for each operating state, and the output of the engine 10 at that time is measured by the measuring device 35. Each condition set in the condition file is set by the condition setting tool 53.

  In order to control the operation state of the engine 10 to each operation state set in the condition file, the automatic measuring device 33 supplies data corresponding to the depression amount of the accelerator pedal to the ECU 30 via the panel checker 34.

  When the engine 10 is controlled to the operation state set through this condition file, the automatic measuring device 33 sets a manual flag in a memory or a register in the ECU 30 via the panel checker 34. This manual flag is a flag for prohibiting the control of the engine 10 by the control map.

  When the engine 10 enters the operation state set through the condition file, the automatic measuring device 33 sets this flag and controls the control parameter of the engine 10 to a value set in the condition file.

  With the control parameters fixedly controlled to predetermined control values based on the engine operating conditions set in the condition file, the fuel consumption, the NOx concentration in the exhaust, the output torque fluctuation amount, etc. The characteristic value is measured by the measuring device 35.

  Specifically, the measuring device 35 includes a fuel consumption meter that measures the amount of fuel supplied to the engine 10, an analyzer that analyzes the NOx concentration in the gas component discharged from the exhaust passage 17 of the engine 10, A torque meter installed between the dynamometers 31 and a torque variometer for calculating the value of the torque meter are included.

  Then, with respect to the fuel consumption, the measurement value obtained by the fuel consumption meter is calculated in the automatic measuring device 33. The NOx concentration is calculated by the automatic measuring device 33 using the concentration calculated by the analyzer as a measured value. Further, the fluctuation amount of the output torque is measured as a value of a torque fluctuation meter, and is calculated by the automatic measuring device 33. Data calculated in the automatic measuring device 33 becomes measurement data.

  The adaptive system further includes a server 40 that holds measurement data for each condition file, an analysis tool 50 that analyzes the measurement data held in the server 40 together with information about each condition file, and an analysis result by the analysis tool 50. Is displayed, a database 52 for storing and holding a part of the analysis result, and an operation unit 60 for operating the analysis tool 50, the condition setting tool 53, and the like.

[Explanation of missing points and additional test points]
FIG. 2 is a diagram for simply explaining missing points.

  In FIG. 2, in order to briefly explain the features of the embodiment of the present invention, an experimental plan in the case of changing two parameters of the variable valve timing VVT and the ignition timing AOP as engine control parameters will be described.

  In order to increase the accuracy of the engine characteristic model, it is preferable to apply a second order model. In this case, three levels of experimental points are set in which the values of the control parameters are changed in three values. Assuming that the center point of each control parameter is “0” and the upper and lower values are “+1” and “−1”, nine experimental points P (−1,1) and P (0,0) as shown in FIG. 1), P (1,1), P (-1,0), P (0,0), P (1,0), P (-1, -1), P (0, -1), P (1, -1) is set. It is assumed that the experimental point P (1,1) is a missing point as a result of the measurement at the above nine experimental points.

  Missing measurement points are when abnormal combustion causes knocking and the engine does not rotate normally, when the engine becomes too hot and exceeds the heat resistance of the engine, or when the fuel is not properly ignited. The case where unburned fuel is discharged and becomes inconvenient for the exhaust gas catalyst is applicable. In addition, the case where the data is measured but the noise is large or the data becomes abnormal due to a communication failure or the like is also applied to the missing point.

  In order to obtain a highly accurate model when the missing point P (1,1) occurs, an experimental plan that determines 9 experimental points from the beginning again after removing the missing point P (1,1). It is conceivable that the measurement is performed again with respect to nine newly determined experimental points.

  However, in such a method, the experimental results of eight experimental points that have been measured so far and have obtained normal data are often wasted.

  Therefore, in the present invention, a secondary model is created using the normal experimental point data and the additional experimental point data obtained so far by setting the additional experimental points while making use of the data measured so far. To do.

  FIG. 3 is a diagram for explaining additional experimental points.

  Referring to FIG. 3, first, a point obtained by dividing the control width of the control parameter into four equal parts is added as a candidate for an additional experiment point. The control width is the difference between the maximum value and the minimum value that can be set for the control parameter.

  The experiment candidate points are thus set, and additional points larger than the number of missing points are selected from the experiment candidate points. For example, the number of additional points is rounded to the number of additional points by rounding off the number of decimal points that is 1.5 times the number of missing points. In the case of FIG. 2, there are one missing point, so in FIG. 3, there are two additional points.

Then, all combinations for selecting two points from the experimental candidate points are evaluated. The evaluation is based on the D optimization plan for the two selected points and the experimental points for which data has been normally obtained so far. Then, a combination of two points with the maximum evaluation is set as an additional experimental point. In the D optimization plan, a predetermined matrix X is created for a combination of two selected points and normal experimental points, and two experimental candidate points such that the determinant | X ^T X | This will be selected, but will be described in detail later. Note that ^XT is a transposed matrix of the matrix X.

  In the case of FIG. 3, the additional points P (0.5, 1) and P (1, 0.5), which are the experimental candidate points closest to the missing point P (1, 1), are selected.

  By doing so, it is possible to make use of the data of 8 experimental points that have already been measured, and to further reduce the number of experimental points to be newly measured to 2. This makes it possible to significantly reduce the time compared to the case where the experimental plan is re-established from the beginning after removing the missing measurement point P (1, 1).

  FIG. 4 is a flowchart for explaining an engine control parameter experiment plan setting method and an engine control parameter optimization process using the method according to the embodiment of the present invention.

  Referring to FIG. 4, the engine control parameter experimental plan setting method sets a first predetermined number of experimental points, each of which is indicated by a combination of a plurality of engine control parameters, necessary for obtaining a response surface. (S1), a step (S2) of inputting experimental result data measured at the set first predetermined number of experimental points, and referring to the experimental result data, data effective for obtaining a response surface is obtained. A step (S3) of extracting the obtained effective experimental points and the missing measurement points for which no valid data was obtained, and testing the missing measurement points when the number of missing measurement points exceeds a predetermined threshold And (S5, S6) setting a second predetermined number of additional experimental points that can be obtained in combination with effective experimental points and then used in combination with effective experimental points.

  Referring to FIGS. 1 and 4, when the process is started, an experiment plan for data measurement is set in condition setting tool 53 in step S <b> 1. The set conditions are transferred to the automatic measurement device 33 as a condition file and stored.

  Subsequently, in step S2, the automatic measuring device 33 controls the panel checker 34 and the dynamo operating panel 32 based on the condition file to cause the engine 10 to perform an operation corresponding to each experimental point. Collect characteristic values as measurement data. Then, the automatic measuring device 33 transfers the collected measurement data to the server 40. Then, the measurement value is input from the server 40 to the analysis tool 50.

  Subsequently, unmeasured processing is performed in step S3. In the unmeasured process, the analysis tool 50 calculates the number of missing points with reference to the measurement data and the condition file. Missing measurement points include a case where a flag indicating that measurement was not possible is written in the measurement data, or data that seems to be noise that would cause the measurement value to exceed a predetermined threshold Is applicable.

  In step S4, it is determined whether or not the number of missing data exceeds a predetermined number, for example, 3. This predetermined number is adjusted according to the number of control parameters and the number of experimental points.

If the number of missing data exceeds the predetermined number in step S4, a normal model cannot be created, and the process proceeds to step S5 where additional planning processing is performed. In the additional planning process, the D optimization plan is used to determine additional points to be added to the experimental points that have been normally measured so far so that the determinant | X ^T X |

  In step S6, an order is set so that the additional experimental points can be efficiently tested. When step S6 is completed, the conditions for the additional points are transferred to the automatic measuring device, and based on this, the panel checker 34 and the dynamo operation panel 32 are controlled, and data corresponding to the additional experimental points is collected by the measuring instrument 35. The The data collected again is transferred from the server 40 to the analysis tool 50 in step S2. In the analysis tool 50, the unmeasured process in step S3 is performed again, and it is determined in step S4 whether or not the number of missing measurements is greater than a predetermined number.

  If it is determined in step S4 that the missing number is equal to or less than the predetermined number, it is a case where a proper model can be obtained only from data of experimental points that are normally measured. In this case, missing value processing is performed in step S7 to determine the current missing number as a missing value, and characteristic value modeling processing is performed in step S8.

  In the modeling process, a model formula that defines the relationship between each control parameter and each characteristic value of the engine 10 is obtained for each operating state based on the measurement result measured normally.

  In step S9, an engine control parameter optimization process is performed using the model formula of the obtained characteristic value. In this optimization process, the adaptive value of the control parameter in each operating state is calculated from the obtained model formula based on the predetermined conditions for each characteristic value of the engine 10 for each operating state. Then, the process ends when the adaptive value of the control parameter is calculated.

[Specific example when there are three control parameters]
Subsequently, as a more specific example, a specific example in which the experiment design setting method according to the embodiment of the present invention is applied to a case where there are three control parameters will be described.

  For example, consider the variable valve timing VVT, the ignition timing AOP, and the rate EGR of mixing exhaust into the intake air as engine control parameters. Further, as an example of the engine output characteristics, the fuel consumption BSFC, the nitrogen oxide concentration NOx, and the torque fluctuation TF are optimized.

  Since engine output characteristics are non-linear with respect to changes in each control parameter, at least three levels of experimental data are required for each control parameter.

  FIG. 5 is a diagram showing experimental points in a general experimental design method.

  As shown in FIG. 5, the conventional experimental design requires 3 to the third power, that is, 27 experimental points.

  However, in recent years, it has been recommended to set a more efficient plan by a central composite plan (CCF) or the like.

  FIG. 6 is a diagram showing experimental points set by the central composite design.

  In the example of FIG. 6, the ignition timing AOP is shown as three experimental values of 15 degrees (° BTDC), 25 degrees, and 35 degrees. In addition, the variable valve timing VVT is shown as set values at three points of crank angles 0 degrees, 20 degrees, and 40 degrees. Further, the exhaust gas ratio EGR with respect to the intake air is shown as a set value of 0%, 10%, and 20%. Fifteen experimental points are selected depending on the combination of the set values of these control parameters.

  In the case of FIG. 5, it was 27 experimental points, but in the case of FIG. 6, it is reduced to 15 experimental points. Therefore, the importance of each measurement value is increased as compared with the case of FIG. However, the center point is measured several times in order to take into account variations. Therefore, the center point is the experiment number NO. 8-NO. 10 corresponds.

  FIG. 7 is a diagram showing an example in which engine output characteristics are measured with respect to control parameters set by the central composite plan of FIG.

  The measurement numbers 1 to 17 in FIG. 1-NO. 17 respectively.

  FIG. 8 is a diagram in which the output characteristics are converted into mathematical formulas, the calculated values are obtained, and the difference from the measured value is shown as the residual.

  In FIG. 8, the fuel consumption BSFC, the nitrogen oxide concentration NOx, and the torque fluctuation TF, which are engine output characteristics, are approximated by a model expression of a quadratic response curved surface. However, since the torque fluctuation TF is a characteristic that does not have a normal distribution, the logarithmic conversion is performed, and then approximated by a model expression of a quadratic response surface and inversely converted.

  Here, the model formula will be explained a little. In the present embodiment, it is an object to obtain a suitable value of the engine control parameter at each point for a plurality of operating conditions (for example, 120 points) determined by the rotational speed and load of the engine 10.

  The calculation of the conformity values at the plurality of points is basically as follows: 1) For each operating state, several control parameter values are set and the above-described characteristic values of the engine 10 are measured; Based on the measurement results, a model equation that defines the relationship between each control parameter and each characteristic value of the engine 10 is obtained for each operating state; 3) For each characteristic value of the engine 10 for each operating state This is performed by a procedure of calculating an appropriate value of a control parameter in each operation state from the obtained model formula based on a preset requirement condition.

The quadratic model equation is assumed as the following equation (1) when the _three control parameters are X ₁ , X ₂ , and X ₃ .
β ₁ X ₁ + β ₂ X ₂ + β ₃ X ₃ + β ₁₂ X ₁ X ₂ + β ₁₃ X ₁ X ₃ + β ₂₃ X ₂ X ₃ + β ₁₁ X ₁ X ₁ + β ₂₂ X ₂ X ₂ + β ₃₃ X ₃ X ₃ + β ₀ = Y ... (1)
By interpolating between the experimental points with the model equation, the state of the response can be understood. That is, if each coefficient β is obtained, a response surface can be obtained.

  A method for calculating each coefficient β from the measurement data will be described.

Expression (1) in a matrix format with each experimental point P as one line is as follows. However, X ₁ (1) is the experimental point NO. The measurement data of X1 in ₁ are shown. For simplification of description, X ₁ (1) × X ₂ (1) and the like are written as X ₁ X ₂ (1).

If X is a matrix and β and y are an array of the number of experimental points, the following equation (2) is obtained.
Xβ = y (2)
Therefore, the coefficient β is obtained by the equation (3).
β = (X ^T X) ⁻¹ X ^T y (3)
Here, in FIG. 8, when the fluctuation range of the fuel consumption BSFC is about 20%, if the residual BSFC is within ± 2%, it is considered that modeling by the formula is performed well. If the residual of torque fluctuation TF is within ± 0.1, the mathematical formula can be used as a model. Further, the residual of the nitrogen oxide concentration NOx is almost zero, so it is well modeled.

  From the above, it is considered that the engine output characteristics BSFC, NOx, TF can be expressed by model equations with respect to changes in engine control parameters.

  FIG. 9 is a diagram showing a response curved surface based on the fuel consumption rate BSFC according to the model formula at the 17 experimental points shown in FIG.

  FIG. 9 shows a response curved surface when the variable valve timing VVT is 0 degree. In FIG. 9, the experiment numbers 1, 2, 3, 4, and 5 in FIG. 8 are indicated by arrows for the corresponding reference.

  Here, a case where experimental data cannot be obtained will be considered. Since it is a measurement under the conditions narrowed down by the experimental design method, it is basically thought that if normal data cannot be obtained at the experimental point, the accuracy of the model obtained by that will be extremely deteriorated, In the missing measurement of 1 to 2 points, even if the data is used as it is, an almost good model can be obtained for portions other than the missing measurement point. Note that the missing point is not a point where the engine is actually operated, so even if the calculated value at the missing point deviates somewhat from the appropriate value, there is no problem in finding a suitable point for actually operating the engine. .

  However, when the number of missing points increases, the difference between the calculated value and the actually measured value in the vicinity of the missing point is greatly different.

  FIG. 10 is a diagram illustrating the positions of missing points in an example where there are four missing points.

  Referring to FIG. Consider a case where four points 1, 5, 11, and 17 are missing points where data cannot be normally obtained. Missing points occur when knocking occurs or when unburned gas is mixed with exhaust due to misfire and burned in the catalyst.

  FIG. 11 is a diagram showing the difference between the calculated value and the actually measured value when there are four missing points.

  In FIG. 11, NO. Instead of using the measurement data of the four missing points of 1, 5, 11, and 17, use the data of other experimental points to approximate the engine output characteristics with a quadratic response surface and output the calculated value Yes.

  The reason why the residual is obtained even at the missing points is that the measured values in FIG. 7 are obtained also at the four missing points. For example, in actual vehicles, it is assumed that mass-produced catalysts with low heat resistance are used, and regular gasoline is usually used as the fuel. The data can be obtained by using a high heat resistance or using high-octane gasoline to prevent knocking.

  In FIG. 11, it can be seen that the residual is particularly large at missing points. However, since the missing point is an area that cannot actually be used, even if the residual at the missing point becomes large, the model is good if the residual at the other operable area is small.

  FIG. 12 is a diagram showing a response surface based on the calculated value of the fuel consumption rate BSFC corresponding to FIG.

  FIG. 12 shows a response curved surface when the variable valve timing VVT is 0 degree. Comparing FIG. 12 and FIG. 9, NO. It can be seen that the calculated value deviates greatly in the vicinity of 1 as compared with the normal case. In this way, the result was greatly different from the normal model. In other words, when the number of missing points increases, the calculated value is far from the normal model in the vicinity of the missing point, so it becomes necessary to increase the number of experimental points and add data.

  Consider the missing point conditions shown in FIG. 10 for increasing experimental points. In FIG. 10, the combination of AOP = 35 and EGR = 20 cannot be measured. Therefore, in order to set the experimental point in the experimental region, the control parameter ranges are set to AOP = 15-30 and EGR = 0-15. It is also possible to re-plan and re-measure the plan. However, all the re-planned experimental points must be measured, and the data of the experimental points measured so far are also wasted and become very inefficient.

  In addition, it is desirable that the control parameter setting value be a value that is easy to experiment. However, there may be a case where the control parameter setting value at the re-planned experimental point becomes a halfway value and the experiment becomes difficult.

  FIG. 13 is a diagram showing additional experimental points when a missing point as shown in FIG. 10 exists.

  As shown in FIG. Five experimental points 21 to 25 are added, and data measured at these experimental points are used in an approximate expression approximated by a quadratic response surface. That is, the necessary additional points NO. Only 21 to 25 are measured to ensure the accuracy of the model formula.

  First, a point obtained by dividing the experimental space into four equal parts for each control parameter is set as an additional experimental candidate point. The points obtained by dividing the experimental space into four parts are often easier to set the control parameter values during the experiment than the points divided into three parts. Here, as a method for selecting an additional experimental point from these additional experimental candidate points, the D maximum drop design is used.

In the equation (3) showing the response surface described above, the accuracy of the coefficient β depends on the stability of (X ^T X) ⁻¹ . This stability can be evaluated by the magnitude of the determinant | X ^T X |. In the D optimization plan, experimental points are determined so as to maximize the size of the determinant | X ^T X |.

In other words, the matrix X is created by adding the number of points + α missing from the experimental candidate points to the matrix X ′ excluding the missing points. In the example of FIG. 13, with respect to the missing point 4, α = 1 and the number of additional points is five. Then, from the result of calculating the determinant | X ^T X |, the maximum combination of additional points is determined. Additional experimental points in FIG. 21 to 25 are obtained in this way.

  I will explain this determinant a little.

Consider a quadratic response surface as a model in a central composite design with three levels of control parameters. For simplicity, the three level values are considered to be dimensionless into three values of “−1”, “0”, and “1”. This is because the magnitude relationship between the calculation results of the determinant | X ^T X |

  FIG. 14 is a diagram illustrating matrices corresponding to the equations (1) and (2) when dimensionless.

  If the average value is the origin, the experimental design of the three parameters is as indicated by X1 to X3 in FIG. Further, the interaction is represented by X1X2, X1X3, X2X3, and the secondary component is represented by X1X1, X2X2, X3X3.

When the determinant | X ^T X | is obtained by setting the matrix shown in FIG. 14 to X, the answer is 53248000.

  FIG. 15 is a diagram illustrating a matrix when there are four missing values.

In FIG. 15, the missing value NO. 1, 5, 11, and 17 are excluded from FIG. In this case, when the determinant | X ^T X | is calculated, the answer is 12288, which is much smaller than that of the original matrix of FIG. Therefore, this determinant is enlarged by adding rows of five effective additional points to the matrix shown in FIG.

  FIG. 16 shows the additional experiment point NO. It is the figure which showed the matrix to which the line corresponding to 21-25 was added.

In the case of the matrix shown in FIG. 16, when the determinant | X ^T X | is calculated, the answer is 404733993.97. In this case, the additional points are based on the number of missing points + 1, and the additional experimental points located on the straight line connecting the adjacent missing values are removed, and the experimental space is divided into four equal parts, that is, the dimensionless value is 0. The analysis tool 50 in FIG. 1 obtains a point that maximizes the determinant on the combination with the additional experimental points in increments of 5 as candidates. In this way, the additional experiment point NO. 21 to 25 are required. The analysis tool 50 can generally use a computer.

  FIG. 17 is a diagram in which the measurement data of the additional experimental points 21 to 25 is displayed in addition to the experimental points excluding the missing measurement points.

  FIG. 18 is a diagram showing the difference between the calculated value and the measured value based on the model equation approximated using the original valid data and the additional point data.

  Also in FIG. 18, each point was estimated from an equation approximated by a quadratic response curved surface. Comparing FIG. 18 with FIG. 8, it can be seen that the calculated values of FIG. 8 obtained without missing measurement and the calculated values of FIG.

  As explained above, when there are missing points like an automobile engine, the accuracy of the model of each characteristic can be efficiently improved by performing few additional experiments while making use of normal data already obtained by experiments. It can be secured. As a result, the time required for adapting the engine control parameters can be shortened.

[Modification of Embodiment]
In the above embodiment, in the experimental design method, when engine breakage occurs when the control is performed at the experimental point, the test condition is set and the test is performed in the additional plan for adding the additional point, and finally obtained. The case where each characteristic is modeled from the measured point data has been described.

  As the characteristic model, a model based on a physical quantity and a model based on a control quantity are considered. For engine control adaptation, a model based on the control amount is required to obtain an optimal solution with the control parameters. However, further model accuracy is required for adaptation considering robustness for quality improvement. The In order to form a model with high accuracy, there is a problem such as collinearity in the controlled variable, so an experiment plan based on a physical quantity is indispensable.

However, the control amount of the engine (Y) are different two or more physical quantity parameters (X _1, X ₂₎ in the case represented by the controlled variable (Y) to the corresponding 2 or more physical quantity parameters (X _1, If the experimental design method is used for the parameter of the physical quantity in the region where X ₂ ) does not exist, the accuracy of the response surface deteriorates.

  FIG. 19 is a diagram for describing a region where the engine control amount does not exist.

Referring to FIG. 19, the vertical axis indicates the overlap amount ol, and the horizontal axis indicates the intake valve closing timing vtcls. The overlap amount ol refers to an angle at which both the intake valve and the exhaust valve are opened, and is represented by a crank angle. When the advance amount of the intake valve is invt and the advance amount of the exhaust valve is exvt, the overlap amount ol is expressed by the following equation (4).
ol = invt + exvt + ofset0 (4)
Ofset0 is the overlap amount when invt = 0 and exvt = 0. At this time, ofset0 can take a negative number, but when ol <0, it is necessary to consider that ofset0 is 0.

That is, the overlap amount ol, which is a control amount, is expressed as a function f with a physical quantity parameter of 2 or more. At this time, it can be expressed as the following equation (5).
ol = f (invt, exvt, ofset0) (5)
Similarly, the intake valve closing timing vtcls, which is a control amount, can be expressed as a function g as in the following equation (6).
vtcls = g (invt, ofset2) (6)
Ofset2 is a constant for calculating vtcls from invt.

  It is assumed that an experimental point by the experimental design method is determined in the area A1 in FIG. However, when the overlap amount ol and the intake valve closing timing vtcls are indicated by related parameters in this way, the experimental point that can be actually tested is indicated by the region A2, and the experimental point that cannot be tested in the region A1. May be included.

  FIG. 20 is a diagram illustrating a state in which additional points are determined with respect to the experimental points determined by the experimental design method.

  Referring to FIG. 20, experimental points P (-1, 1), P (0, 1), P (1, 1), P (-1, 0), P (0, 0), P (1, 0), P (-1, -1), P (0, -1), and P (1, -1) are the initial experimental design points set by the central composite design in the area A1 in FIG. is there. Of these initial experimental design points, the experimental points that protrude from the area A2 in FIG. 19 are missing points. In FIG. 20, two points, experimental points P (1, 1) and P (-1, -1), are missing points.

  Even when a missing point occurs in such a case, a point obtained by dividing the control width of the control parameter into four equal parts as in the case described with reference to FIG. In this case, double the number of missing points as the additional points. Then, additional experimental points are determined by the D optimization plan. FIG. 20 shows that four additional experimental points P (−0.5, 0.5), P (0.5, 0.5), P (−0.5, −0.5), and P (0.5, −0.5) are defined.

  FIG. 21 is a flowchart for explaining an engine control parameter experiment plan setting method and an engine control parameter optimization process using the same according to a modification of the embodiment of the present invention.

Referring to FIGS. 1 and 21, when the process is started, a physical model of engine characteristics is created in step S11. Assuming that engine characteristic values such as TQ and NOx are Y1 and Y2, the engine characteristic values are represented by the following equations using model functions F1 and F2 such as a secondary response model. Note that st represents the ignition timing.
Y1 = F1 (ol, vtcl, st,...) (7)
Y2 = F2 (ol, vtcl, st,...) (8)
Subsequently, in step S12, an experiment plan for measuring data with physical quantities is set. In step S13, the relational expression between the control amount and the physical quantity is set as described in the equations (4) to (6). Step S13 may be performed before step S11.

Then, as described with reference to FIG. 20, in step S14, processing for excluding missing measurement points from the experimental points is performed in consideration of the controllable setting range, and in step S15, the experimental points excluding missing measurement points are processed. It is determined whether or not the model formula can be calculated. If the model formula cannot be calculated in step S15, the process proceeds to step S16, where the experimental points that cannot be measured are missing, and points twice the number of missing points are added as additional experimental points by the additional plan (D optimization plan). Then, the determination in step S15 is performed again. In the D optimization plan, an additional point to be added to the experimental point that is not a missing measurement set so far so that the determinant | X ^T X | is maximized is determined.

  When the model formula can be calculated in step S15, the process proceeds to step S17, and data is measured based on the experiment plan. At the time of measurement, the measurement is performed by calculating back the control amount based on the physical quantity.

  Subsequently, in step S18, the physical value prediction formula calculation process for the characteristic value is performed. In the physical model prediction formula calculation process, a model formula that defines the relationship between each physical quantity control parameter and each characteristic value of the engine 10 is obtained for each operating state based on the measurement result measured normally. In this process, a secondary response surface based on physical quantities is obtained by a statistical method. Although a quadratic response surface is used as a model, if a quadratic item is not significant (determined using a test method), the prediction model is improved by deleting the item.

  In step S19, the physical quantity is converted into a control quantity, and a control quantity characteristic prediction formula calculation process is performed.

  Finally, in step S20, an engine control parameter optimization process is performed using the obtained characteristic value model expression. In this optimization processing, each operation state is determined from the obtained model formula in consideration of the preset conditions for each characteristic value of the engine 10 for each operating state, that is, the limit value and trade-off of each characteristic. The control parameter conformity value in the state is calculated. The process ends when the adaptive value of the control parameter is calculated.

  In the modification of the embodiment, an experimental design is made based on the physical quantity and the control amount is calculated back, so that a highly accurate secondary response surface can be obtained and it is easy to remove the uncontrollable range, so it is compatible. Efficiency is promoted.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

1 is a block diagram showing a configuration of an engine control parameter adaptation system to which an engine control parameter experiment plan setting method according to an embodiment of the present invention is applied. FIG. It is a figure for demonstrating a missing point simply. It is a figure for demonstrating an additional experimental point. It is a flowchart for demonstrating the engine control parameter experiment plan setting method and engine control parameter optimization process using the same according to the embodiment of the present invention. It is the figure which showed the experimental point in a general experiment design method. It is the figure which showed the experimental point set by central composite design. It is a figure which shows an example which measured the engine output characteristic with respect to the control parameter set by the center compound plan of FIG. It is the figure which calculated | required the output characteristic, calculated | required the calculated value, and showed the difference with the measured value as a residual. It is the figure which showed the response curved surface by the model type | formula about the fuel consumption rate BSFC at the experimental point of 17 shown by FIG. It is the figure which showed the position of the missing point about the example when there are 4 missing points. It is the figure which showed the difference of a calculated value and measurement value when four missing measurement points exist. It is the figure which showed the response curved surface based on the calculated value of the fuel consumption rate BSFC corresponding to FIG. It is the figure which showed the additional experiment point when the missing point as shown in FIG. 10 exists. It is the figure which showed the matrix corresponding to Formula (1) and (2) at the time of making it dimensionless. It is the figure which showed the matrix when the missing value of 4 points | pieces exists. Additional experiment point NO. It is the figure which showed the matrix to which the line corresponding to 21-25 was added. It is the figure which displayed the measurement data of the additional experimental points 21-25 in addition to the experimental point except a missing measurement point. It is the figure which showed the difference of the calculated value and measured value by the model formula approximated using the original effective data and the data of an additional point. It is a figure for demonstrating the area | region where an engine control amount does not exist. It is a figure which shows the state which defined the additional point with respect to the experimental point defined by the experiment design method. It is a flowchart for demonstrating the experiment plan setting method of the engine control parameter which concerns on the modification of embodiment of this invention, and the optimization process of an engine control parameter using the same. It is a figure for demonstrating generally the adaptive value of the control parameter of an engine.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 Engine, 11 Cylinder, 12 Piston, 13 Combustion chamber, 14 Injector, 15 Spark plug, 16 Intake passage, 17 Exhaust passage, 18 Intake valve, 19 Exhaust valve, 20 Variable valve timing mechanism, 21 Electronic control throttle, 22 EGR passage , 23 EGR valve, 24 output shaft, 25 rotational speed sensor, 26 water temperature sensor, 31 dynamometer, 32 dynamo operation panel, 33 automatic measuring device, 34 panel checker, 35 measuring instrument, 40 server, 50 analysis tool, 51 display , 52 database, 53 condition setting tool, 60 operation unit.

Claims (11)

Setting experimental points each indicated by a combination of a plurality of engine control parameters by a first predetermined number necessary to obtain a response surface;
Inputting experimental result data measured at the set first predetermined number of experimental points;
Referring to the experimental result data, extracting effective experimental points where data effective to obtain the response surface and missing points where the effective data could not be obtained;
When the number of missing measurement points exceeds a predetermined threshold, the response curved surface can be obtained by using the missing measurement points in combination with the effective experimental points after eliminating the missing measurement points from the experimental points. A method of setting an engine control parameter experiment plan, comprising: setting a second predetermined number of additional experiment points. The step of setting the additional experimental point includes:
When a matrix generated using the plurality of engine control parameters corresponding to each of the effective experimental points and the plurality of engine control parameters corresponding to the second predetermined number of additional experimental points is X, a determinant | Determining a combination of additional experimental points where X ^T X |
The engine control parameter experiment plan setting method according to claim 1.   The engine control parameter experiment plan setting method according to claim 1, wherein the first predetermined number of experiment points is determined based on a central composite plan. The experimental candidate points when setting the first predetermined number of experimental points are the maximum value and the minimum value of the control range of the engine control parameter, and the three selected values of the intermediate value between the maximum value and the minimum value. Determined by the combination of selecting any one of each engine control parameter,
The additional experimental point is any one of five selected values obtained by adding a point between the maximum value and the intermediate value and a point between the minimum value and the intermediate value to the selected value of the three points. 4. The engine control parameter experimental plan setting method according to claim 3, wherein the engine control parameter experimental plan setting method is selected from the combinations in which values are selected for each engine control parameter from which the missing measurement points are removed. In the step of setting the first predetermined number necessary for obtaining the response surface, when a certain engine control parameter can be expressed by two or more other parameters, the engine control parameter at the experimental point is determined by the other two or more parameters. A step of setting as an experimental point only when it is in the operating range;
The engine control parameter experimental plan setting method according to claim 1, further comprising a step of setting a missing point when the engine control parameter at the experimental point is outside the operating range of the other two or more parameters.   The program which makes a computer perform the experiment plan setting method of the engine control parameter of any one of Claims 1-5.   The computer-readable recording medium which recorded the program which makes a computer perform the experiment plan setting method of the engine control parameter of any one of Claims 1-5. A first setting unit that sets only a first predetermined number of experimental points each indicated by a combination of a plurality of engine control parameters necessary to obtain a response surface;
An input unit for inputting experimental result data measured at the set first predetermined number of experimental points;
With reference to the experimental result data, an extraction unit that extracts effective experimental points where data effective for obtaining the response surface was obtained and missing points where the effective data was not obtained,
When the number of missing measurement points exceeds a predetermined threshold, the response curved surface can be obtained by using the missing measurement points in combination with the effective experimental points after eliminating the missing measurement points from the experimental points. An engine control parameter experiment plan setting apparatus comprising: a second setting unit configured to set a second predetermined number of additional experiment points. The second setting unit includes:
When a matrix generated using the plurality of engine control parameters corresponding to each of the effective experimental points and the plurality of engine control parameters corresponding to the second predetermined number of additional experimental points is X, a determinant | 9. The engine control parameter experiment plan setting device according to claim 8, wherein a combination of additional experiment points at which X ^T X |   9. The engine control parameter experiment plan setting device according to claim 8, wherein the first predetermined number of experiment points is determined based on a central composite plan. The experimental candidate points when setting the first predetermined number of experimental points are the maximum value and the minimum value of the control range of the engine control parameter, and the three selected values of the intermediate value between the maximum value and the minimum value. Determined by the combination of selecting any one of each engine control parameter,
The additional experimental point is any one of five selected values obtained by adding a point between the maximum value and the intermediate value and a point between the minimum value and the intermediate value to the selected value of the three points. 11. The engine control parameter experimental plan setting device according to claim 10, wherein the engine control parameter experimental plan setting device is selected from combinations in which values are selected for each engine control parameter from which the missing measurement points are removed.

Images