JP2005042656A - Method for adapting control parameter and device for adapting control parameter of vehicular engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To further efficiently adapt control parameter satisfying total engine characteristic requirement in a predetermined travelling mode. <P>SOLUTION: A simulation system 70 calculates total engine characteristic in the travelling mode while simulating behavior of a vehicular engine during vehicle travelling in the predetermined travel mode based on a model formula indicating correlation of a engine characteristic value and control parameters determined through measurement of the engine characteristic value based on design of experiment. An operation map with considering a travelling condition of a vehicle is adapted by evaluating whether the total engine characteristic requirement in the travelling mode is satisfied based on the calculation result and renewing a map value of the operation map of control parameters according to the evaluation result. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method for adapting control parameters of an in-vehicle engine and an apparatus for performing the adaptation.

  As is well known, in the control of an in-vehicle engine, various control parameters such as a fuel injection timing and an ignition timing are determined according to an engine operating state such as an engine speed and an engine load. Each control parameter in each operating state is preliminarily adapted so that various engine characteristics such as exhaust emission characteristics, ignition characteristics, and fuel consumption characteristics satisfy the requirements.

  The adaptation of the control parameters is performed by repeated trial and error on the engine bench. That is, the output shaft of the in-vehicle engine and the dynamometer are connected by a rotational drive shaft, and the load torque of the in-vehicle engine is absorbed as a test torque by the dynamometer, so that the in-vehicle engine is mounted and operated in the vehicle. Simulate. Then, various engine characteristic values such as nitrogen oxide emission amount and fuel consumption amount are measured while adjusting the control parameter in each operation state, and the optimum value of the control parameter is acquired as a suitable value. The adaptation of such control parameters requires trial and error and a great deal of time.

  Therefore, conventionally, a matching method as shown in Patent Documents 1 and 2 has been proposed. In these adaptation methods, first, each engine characteristic value is measured while changing the control parameter in each operation state of the in-vehicle engine, and a model of the control parameter and the engine characteristic value in each operation state based on the measurement result. Seeking an expression. Then, using the model formula, an appropriate value of the control parameter that satisfies the requirements of the engine characteristics in each operation state is determined. For example, in each operating state, the set value of the control parameter that minimizes the amount of fuel consumption is determined as an appropriate value within a range where the nitrogen oxide emission amount and the torque fluctuation amount are below the allowable level. As a result, trial and error on the benchmark can be reduced, and the efficiency of the adaptation work can be improved.

  By the way, the final evaluation of the engine characteristics of the in-vehicle engine is performed based on the total engine characteristics in a specific travel mode such as the 10-15 mode. However, it is difficult to set in advance the required conditions for the engine characteristic values in each driving state so as to completely satisfy the total engine characteristic requirements in such travel modes. For this reason, in the conventional adaptation method, an adaptation value that satisfies such total engine characteristic requirements may not always be acquired from the beginning.

Therefore, in the conventional adaptation method, after adapting the engine alone based on the model formula, the operating state of the engine in the travel mode is reproduced on the engine bench, and the total engine characteristics are measured. The validity of the obtained conformity value is evaluated. As a result of the evaluation, when the total engine characteristic requirement is not satisfied, the requirement condition for the engine characteristic in each operation state is reviewed, and the conforming value is recalculated based on the revised requirement condition. The conformity value is finally determined by repeating such review of the requirement, recalculation of the conformance value, and evaluation until the total engine characteristic requirement is satisfied. As described above, even with the above-described conventional adaptation method, the repetition of trial and error on the engine bench has not been completely eliminated, and there is still room for improvement in the efficiency of adaptation.
JP 2000-248991 A JP 2001-005038 A

  The present invention has been made in view of the above circumstances, and an object of the present invention is to control an in-vehicle engine that can more efficiently perform control parameter matching that satisfies a total engine characteristic requirement in a predetermined traveling mode. A parameter adaptation method and a control parameter adaptation apparatus are provided.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
According to the first aspect of the present invention, in-vehicle engine control parameter adaptation that adapts the control parameter in each operating state so that each of the plurality of engine characteristics of the in-vehicle engine satisfies a total engine characteristic requirement in a predetermined traveling mode. In the method, an engine characteristic value is measured by changing the control parameter in each operation state, and a model equation representing a correlation between the control parameter and the engine characteristic value in each operation state based on the measurement value A step of calculating, a step of setting the control parameter of each driving state based on the calculated model formula, and each of when the in-vehicle engine is operated in the traveling mode based on the set control parameter Calculate the engine characteristic value at the time based on the model formula, A step of evaluating whether or not the total engine characteristic requirement is satisfied based on the output result, and according to a result of the evaluation, from the target value of the engine characteristic value that does not satisfy the total engine characteristic requirement Updating the control parameter setting mode based on the model formula so as to reduce the degree of deviation, and until the total engine characteristic request is satisfied for all of the plurality of engine characteristics in the evaluation The gist of the invention is to repeatedly execute the setting of the control parameter, the update of the setting mode, and the evaluation to obtain the adaptive value of the control parameter in each operation state.

  In the above method, a model formula representing the correlation between the control parameter and the engine characteristic value in each driving state is calculated based on the measurement result of the engine characteristic value of the in-vehicle engine. And based on the model formula, the control parameter of each driving | running state of a vehicle-mounted engine is set. Since the control parameter is set based on a model expression that represents the correlation between the control parameter and the engine characteristic value, the control parameter is set so that the engine characteristic value in each engine operating state is a desired value. Can be set.

  When the control parameters for each operating state are set in this way, the engine characteristic value at each time when the vehicle-mounted engine is operated in a predetermined traveling mode based on the set control parameters is calculated based on the model equation. . If the engine characteristic value at each time in the traveling mode is obtained in this way, the total engine characteristics in the traveling mode can also be obtained. Therefore, in the above method, based on the calculated engine characteristics at each time, whether or not the total engine characteristic requirement is satisfied, that is, the validity of the control parameter in each operating state set above is evaluated. Done.

  Furthermore, in the above method, the control parameter setting mode based on the model formula is updated so that the degree of deviation from the target value of the engine characteristic value that does not satisfy the total engine characteristic requirement is reduced according to the result of the evaluation. Is done. If the control parameters are set on the basis of the updated setting mode in this way, at least for engine characteristics that do not satisfy the total engine characteristic requirements, values that will change the engine characteristic value to the desired side in each operating state Is set as the control parameter of the operating state.

  In the evaluation, setting of the control parameter based on the model formula, updating of the setting mode, and evaluation are repeatedly performed until all items of the total engine characteristic request are satisfied. The control parameter in each operation state that is finally set through the repetition of such processing is a value that satisfies the total engine characteristic requirement in a predetermined travel mode. For this reason, the adaptive values of the control parameters that satisfy the total engine characteristic requirements in the predetermined traveling mode can be obtained easily and quickly without repeating trial and error on the engine bench.

  Therefore, according to the above method, it is possible to more efficiently adapt the control parameters that satisfy the total engine characteristic requirement in the predetermined traveling mode. In addition, since the model formula used for adaptation of such control parameters is obtained from the measured value of the engine characteristic value in the in-vehicle engine, a model formula with high accuracy and reliability can be obtained, and the reliability of the calibration value is It will also be improved.

The invention described in claim 2 is the control parameter adaptation method for a vehicle-mounted engine of claim 1, the setting of the control parameters based on the model equation, A _1, A _2, ..., each of A _n engine when the characteristic values were f _m the (a _m) an evaluation function that represents the deviation degree from the target value for any of the engine characteristic values a _m,
F (A ₁ , A ₂ ,..., A _n ) = f ₁ (A ₁ ) + f ₂ (A ₂ ) +... + F _n (A _n )
Is calculated by calculating the control parameter in each operating state in which the value of the comprehensive evaluation function represented by the minimum is expressed, and the update of the control parameter is performed by the comprehensive evaluation of the engine characteristic value that does not satisfy the requirements in the evaluation The gist is that the evaluation is performed by modifying the evaluation function of each engine characteristic value in accordance with the result of the evaluation so that the degree of reflection on the function value becomes larger.

  In the above method, the control parameter for each operating state is set so that the value of the comprehensive evaluation function obtained by adding the evaluation function representing the degree of deviation from the target value for a plurality of engine characteristic values is minimized. When updating the control parameter setting mode according to the evaluation of the total engine characteristic request in the travel mode described above, the degree of reflection of the engine characteristic value related to the item that has not been achieved in the evaluation with respect to the value of the comprehensive evaluation function The evaluation function of each engine characteristic value is corrected so that becomes larger.

  When the degree of reflection of the engine characteristic value relating to the item that has not been achieved in this way with respect to the value of the overall evaluation function is increased, the engine characteristic value in each operating state is further increased in order to minimize the value of the overall evaluation function. It is necessary to change the control parameter of each operation state to a desirable side, that is, a side that easily satisfies the above total engine characteristic requirement. Therefore, according to the above method, it is possible to more accurately update the control parameter to the side where the unachieved item of the total engine characteristic requirement is achieved according to the result of the evaluation. In addition, since the procedure for updating the control parameter according to the evaluation result is established, the processing can be easily automated.

The invention described in claim 3 is the control parameter adaptation method for a vehicle-mounted engine of claim 2, T _m was and the target value for any of the engine characteristic values A _m, k _m the engine characteristic values A _m the when the weighting coefficients of the evaluation function f _m of the engine characteristic values a _m (a _m) is
_{f m (A m) = k} m × | A m -T m |
In the modification of the evaluation function, the ratio of the weight coefficient of the engine characteristic value not satisfying the total engine characteristic request to the weight coefficient of the engine characteristic value satisfying the total engine characteristic request is increased. In this way, the gist is that it is performed by correcting the weighting coefficient of each engine characteristic value.

  In the above method, the value obtained by multiplying the absolute value of the difference from the target value of the engine characteristic value by the weighting coefficient is set as the evaluation function of each engine characteristic value. As a result of the above evaluation, the weighting factor of the evaluation function of the engine characteristic value related to the item for which the total engine characteristic requirement has not been achieved is such that the ratio of the achieved engine characteristic value to the weighting factor of the evaluation function becomes larger. Will be corrected. For engine characteristics with increased weighting factor ratios, if the control parameters are set so as to minimize the value of the overall evaluation function, it is changed to a more desirable side, that is, a side that easily satisfies the above total engine characteristic requirements. It becomes like this. Incidentally, such a change in the ratio can be performed by increasing the weighting factor related to the engine characteristics that have not yet been required, by decreasing the weighting coefficient related to the engine characteristics that have been achieved, or by a combination thereof.

According to such a method, it is possible to update the control parameters accurately according to the result of the evaluation with a relatively simple function, and it becomes easier to realize the automation of the processing.
According to a fourth aspect of the present invention, in the control parameter adaptation method for an in-vehicle engine according to any one of the first to third aspects, the measurement of the engine characteristic value is performed by determining a measurement point based on an experimental design method. The gist of what is done.

  In the above method, by performing measurement based on the experimental design method, a highly accurate model expression is obtained by measurement at a limited measurement point. Therefore, it is possible to greatly reduce the man-hours and time for measurement related to the calculation while ensuring the accuracy and reliability of the calculated model formula.

  Furthermore, the invention according to claim 5 is a control parameter for an in-vehicle engine that adapts a control parameter in each operation state so that each of a plurality of engine characteristics of the in-vehicle engine satisfies a total engine characteristic requirement in a predetermined traveling mode. In the adapting device, a model formula representing a correlation between each engine characteristic value in each driving state and the control parameter is registered and stored, and a model formula storage means that is stored, and the control parameter in each driving state is based on the model formula The engine characteristic value at each time when the vehicle-mounted engine is operated in the travel mode based on the control parameter based on the control parameter is calculated based on the model formula and the engine characteristic at the calculated time Based on the values, the total engine characteristic requirements are met respectively. A deviation from the target value of the engine characteristic value not satisfying the engine characteristic requirement when at least one of the total engine characteristic requirement is not satisfied in the evaluation means for evaluating whether or not the evaluation means evaluates An update means for updating the setting mode of the control parameter based on the model formula by the setting means so that the degree is reduced, and the control of the setting means when the setting mode is updated by the updating means Re-execution means for re-execution of parameter setting and evaluation of the evaluation means, and when the total engine characteristic request is satisfied for all of the plurality of engine characteristics in the evaluation of the evaluation means, the control at that time The gist of the present invention is to include a determining unit that determines a parameter as a conforming value.

  In the above configuration, the control parameter in each operating state is set by the setting unit based on the model formula registered in the model formula storage unit. Since the setting of the control parameter by the setting means is performed based on a model expression representing the correlation between the control parameter and the engine characteristic value, control is performed so that the engine characteristic value in each engine operating state is set to a desired value. Parameters can be set.

  Now, when the control parameters in each driving state are set and the driving state of the vehicle-mounted engine at each time in the travel mode is obtained, the engine characteristic value at each time in the travel mode can be obtained from the model equation. . Therefore, in the above configuration, after setting the control parameter, the evaluation unit calculates the engine characteristic value at each time when the vehicle-mounted engine is operated in a predetermined traveling mode based on the set control parameter based on the model formula. Based on the calculation result, it is evaluated whether or not the total engine characteristic requirement of the travel mode is satisfied.

  Thereafter, if there is an item for which the total engine characteristic request has not yet been achieved, the setting means causes the updating means to reduce the degree of deviation from the target value of the engine characteristic value for each operating state related to the unachieved item. The control parameter setting mode based on the model formula is updated. Then, the processing of the setting unit and the evaluation unit is re-executed by the re-execution unit. At the time of resetting the control parameters at this time, at least for the engine characteristics related to the item that has not been required, the control parameters are set such that the engine characteristic value is changed to a desirable side in each operation state. . Then, the setting of the control parameter in the setting means and the evaluation by the evaluation means are repeatedly executed until all items of the total engine characteristic request are satisfied. When all the items of the total engine characteristic request are satisfied in the above evaluation, the control parameter of each operating state at that time is determined as the appropriate value by the determining means.

  As described above, in the above configuration, the adaptive value of the control parameter that satisfies the total engine characteristic requirement in the predetermined traveling mode can be easily and quickly acquired without repeating trial and error on the engine bench. Therefore, according to the above configuration, it is possible to more efficiently adapt the control parameters that satisfy the total engine characteristic requirement in the predetermined traveling mode.

The invention described in claim 6 is the control parameter adjustment unit for a vehicle-mounted engine of claim 5, wherein the setting means, A _1, A _2, ..., a A _n and each engine characteristic values, f _m ( when the a _m) an evaluation function that represents the deviation degree from the target value for any of the engine characteristic values a _m,
F (A ₁ , A ₂ ,..., A _n ) = f ₁ (A ₁ ) + f ₂ (A ₂ ) +... + F _n (A _n )
The control parameter for each operating state is set based on the model formula so that the value of the overall evaluation function represented by the formula is minimized, and the updating means is an engine characteristic that does not satisfy the total engine characteristic requirement. The gist is to update the setting mode by modifying the evaluation function of each engine characteristic value so that the reflection degree of the value with respect to the value of the comprehensive evaluation function becomes larger.

  In the above configuration, the control parameter of each operating state is set so that the value of the comprehensive evaluation function obtained by adding the evaluation function representing the degree of deviation from the target value for the plurality of engine characteristic values is minimized by the setting means. Is set. Depending on the updating means, the evaluation function of each engine characteristic value is corrected so that the degree of reflection of the engine characteristic value related to the item for which the total engine characteristic request has not been met with respect to the value of the comprehensive evaluation function becomes larger. Is called.

  When the degree of reflection of the engine characteristic value relating to the item that has not been achieved in this way with respect to the value of the overall evaluation function is increased, the engine characteristic value in each operating state is further increased in order to minimize the value of the overall evaluation function. It is necessary to change the control parameter of each operation state to a desirable side, that is, a side that easily satisfies the above total engine characteristic requirement. Therefore, according to the above configuration, it is possible to more accurately update the control parameter to the side where the unachieved item of the total engine characteristic request is achieved according to the result of the evaluation. In addition, since the procedure for updating the control parameter according to the evaluation result is established, the processing can be easily automated.

The invention described in claim 7 is the control parameter adjustment unit for a vehicle-mounted engine of claim 6, T _m was and the target value for any of the engine characteristic values A _m, k _m the engine characteristic values A _m the when the weighting coefficients of the evaluation function f _m of the engine characteristic values a _m (a _m) is
_{f m (A m) = k} m × | A m -T m |
The modification of the evaluation function increases the ratio of the weight coefficient of the engine characteristic value not satisfying the total engine characteristic requirement to the weight coefficient of the engine characteristic value satisfying the total engine characteristic requirement. The gist of this is through the modification of the weight coefficient of each engine characteristic value.

  In the above configuration, the value obtained by multiplying the absolute value of the difference from the target value of the engine characteristic value by the weight coefficient is set as the evaluation function of each engine characteristic value. As a result of the above evaluation, the weighting factor of the evaluation function of the engine characteristic value related to the item for which the total engine characteristic requirement has not been achieved is such that the ratio of the achieved engine characteristic value to the weighting factor of the evaluation function becomes larger. Will be corrected. For engine characteristics with increased weighting factor ratios, if the control parameters are set so as to minimize the value of the overall evaluation function, it is changed to a more desirable side, that is, a side that easily satisfies the above total engine characteristic requirements. It becomes like this. Incidentally, such a change in the ratio can be performed by increasing the weighting factor related to the engine characteristics that have not yet been required, by decreasing the weighting coefficient related to the engine characteristics that have been achieved, or by a combination thereof. According to such a configuration, it is possible to update the control parameters accurately according to the result of the evaluation with a relatively simple function, and it becomes even easier to realize the automation of the processing.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In the present embodiment, a case will be described as an example where the present invention is applied to adaptation of the calculation map of the control parameters of the in-vehicle engine 10 shown in FIG. The in-vehicle engine 10 is configured as an in-cylinder injection type gasoline engine.

  As shown in FIG. 1, the vehicle-mounted engine 10 includes an injector 14 that directly injects fuel into a combustion chamber 13 defined by a cylinder 11 and a piston 12. A spark plug 15 is disposed above the combustion chamber 13.

  Air is sucked into the combustion chamber 13 through the intake passage 16, and the air forms an air-fuel mixture together with the fuel injected from the injector 14. The air-fuel mixture is ignited by the spark plug 15 and burned, and the combustion gas generated by the combustion is discharged from the combustion chamber 13 to the exhaust passage 17 as exhaust. The timings of intake of air from the intake passage 16 and discharge of exhaust to the exhaust passage 17 are set by opening timings of the intake valve 18 and the exhaust valve 19, respectively. In the case of this in-vehicle engine 10, the opening timing of the intake valve 18 is configured to be variably set by a variable valve timing mechanism (VVT) 20.

  The amount of air taken into the combustion chamber 13 of the in-vehicle engine 10 is adjusted by an electronically controlled throttle valve 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 opening amount of the EGR valve 23.

  Such control of the in-vehicle engine 10 is performed by an electronic control unit (ECU) 30. Information from various sensors that measure the operating state of the in-vehicle engine 10 such as the water temperature sensor 26 and the rotation speed sensor 25 provided in the vicinity of the output shaft 24 of the in-vehicle engine 10 is input to the ECU 30 as measurement information. The ECU 30 controls the opening degree of the throttle valve 21, the advance amount control of the opening timing of the intake valve 18 of the VVT 20, the opening amount control of the EGR valve 23, the ignition of the spark plug 15 based on the input measurement information. Various controls such as timing control and fuel injection timing control of the injector 14 are executed.

  These controls are based on various control parameters such as the throttle opening tang, the VVT advance amount vvt, the EGR amount egr, the ignition timing aop, and the fuel injection timing Ainj based on the operating state of the in-vehicle engine 10 grasped based on the measurement information. This is done through the calculation of The calculation of each control parameter is performed using each calculation map stored in the ECU 30 in advance. In the calculation map of each control parameter, the operation state of the in-vehicle engine 10 is defined by the engine rotation speed ne and the engine load kl, and each of the engine rotation speed ne and the engine load kl in each of a predetermined number of different operation states. Control parameter conformity values are stored respectively.

FIG. 2 shows an example of the calculation map of the VVT advance amount vvt. The calculation map shown in the figure shows a VVT advance value adaptation value vvt in each of 120 operating states with respect to _six engine loads kl _{1 to} kl ₆ and 20 engine rotation speeds ne _{1 to} ne _20. (1,1) to vvt (6,120) are stored. In addition, the calculation maps of other control parameters also store the appropriate values of the respective control parameters in each of the 120 operating states.

  In the following, description will be made on the adaptation mode of the calculation map of each control parameter in this embodiment, that is, the setting mode of the adaptation value of the control parameter for each of the 120 operating states stored in the calculation map. Each adaptive value is set so as to satisfy a total engine characteristic requirement in a predetermined traveling mode for each engine characteristic such as fuel consumption characteristics, nitrogen oxide emission characteristics, and torque fluctuation characteristics.

FIG. 3 shows the calculation map adaptation procedure in this embodiment.
When adapting the calculation map, first, in step 100 of FIG. 3, each operation state is measured through measurement of engine characteristic values on the engine bench, that is, fuel consumption amount BSFC, nitrogen oxide emission amount NOx, and torque fluctuation amount TF. A model expression representing the correlation between each control parameter and each engine characteristic value is calculated. The engine characteristic values here are measured by selecting measurement points based on the experimental design method so that the number of measurement points necessary for creating the model formula is reduced as much as possible.

  Subsequently, in step 200, based on the calculated model formula, an optimum value for each driving state of the vehicle-mounted engine 10 alone is obtained. The calculation of the optimum value here is performed so as to satisfy the requirements of the engine characteristics in each of the above operating states.

  In step 300, the model formula and the optimum value calculated by the above procedure are incorporated into a simulation system that simulates the behavior of the in-vehicle engine 10 in the vehicle traveling state in the traveling mode.

In the next step 400, the specifications of the vehicle on which the in-vehicle engine 10 to be matched is mounted and the travel mode are set.
In subsequent step 500, a calculation map of each control parameter used for evaluation of the in-vehicle engine 10 in the simulation system is set. When the process proceeds to step 500 for the first time in the current adaptation, an arithmetic map is set based on the optimum value obtained in step 200.

  In step 600, estimation of total engine characteristics when the vehicle-mounted engine 10 is operated in the travel mode based on the calculated optimum value is performed in the simulation system. In the subsequent step 700, it is evaluated whether or not the total engine characteristic satisfies the requirement based on the estimation result. The calculation of the total engine characteristic related to this evaluation is performed based on the calculated model formula.

  Here, when it is evaluated that the total engine characteristic requirement is not satisfied, in step 800, each control parameter for each operating state is updated. Based on the updated control parameter, the calculation map in step 500 is reset, and the evaluation in step 600 is performed again.

  On the other hand, when it is evaluated in step 600 that all of the total engine characteristic requirements are satisfied, the set value of the calculation map at that time is determined as the appropriate value. As described above, the calculation map of each control parameter is adapted.

(Model formula and optimum value settings)
First, details of the calculation of the model formula in step 100 and the setting of the optimum value in step 200 will be described. The model formula of each engine characteristic value calculated here is defined by the following formula.
- the fuel consumption of a model equation _{BSFC = a 10 + a 11 ×} tang + a 12 × aop + a 13 × Ainj
_{+ A 14 × egr + a 15} × vvt + a 16 × aop × aop
+ A ₁₇ × egr × aop + a ₁₈ × vvt × aop
+ A ₁₉ × egr × vvt + ...
・ NOX = a ₂₀ + a ₂₁ × tang + a ₂₂ × aop + a ₂₃ × Ainj
+ A ₂₄ * egr + a ₂₅ * vvt + a ₂₆ * aop * aop
_{+ A 27 × egr × egr +} a 28 × vvt × vvt
+ A ₂₉ × vvt × tang +…
Torque variation of the model equation _{TF = a 30 + a 31 ×} tang + a 32 × aop + a 33 × Ainj
+ A ₃₄ * egr + a ₃₅ * vvt + a ₃₆ * aop * aop
+ A ₃₇ × Ainj × Ainj + a ₃₈ × vvt × aop
+ A ₃₉ * vvt * tang + ...
Each of the above-mentioned second-order model formulas has an empirical influence on each engine characteristic from the first-order and second-order terms for each control parameter and the term indicating any two interactions of the control parameters. This is a polynomial obtained by removing terms that are considered to be low.

And in order to obtain | require these model formulas about each of the said 120 driving | running states, several control parameter values are set in each driving | running state, and each engine characteristic value of the vehicle-mounted engine 10 is measured. Regarding this measurement, the constant terms (a ₁₀ , a ₂₀ , a ₃₀ ) of the model formula and the coefficients (a ₁₁ , a ₁₂ ,..., A ₂₁ , a ₂₂ ,. _In calculating ₃₁ , a ₃₂ ,..., Each measurement point is set based on an experimental design method in order to obtain the highest accuracy at the minimum measurement point.

  Specifically, for each of the 120 operating states, each control parameter is set at three points: a center point and values above and below it. Then, the center point is “0”, and the upper and lower values are “+1” and “−1”, and 29 points are measured using the orthogonal table illustrated in FIG. This orthogonal table is set for optimizing information obtained from the measurement while reducing the number of measurement points for each model expression including the interaction term of the above aspect. In FIG. 4, the center point is measured three times, the first (first point), middle (12th point), and last (29th point) of the measurement. This is a measure to eliminate the influence of variation.

  Also, a swing width is set in advance for each control parameter corresponding to “+1” and “−1” in the orthogonal table. For example, assuming that the center point of the ignition timing aop is a point where the advance amount is “30 degrees” from the compression top dead center and the amplitude is “4 degrees”, the value of the control parameter used for measurement is “ Corresponding to “0”, “+1”, and “−1”, they are “30 degrees”, “34 degrees”, and “26 degrees”, respectively.

  By the way, in calculating each model formula, in order to increase the reliability, it is desirable to set the measurement point in the vicinity of the matching point, in other words, to set the center point to a value close to the matching value in advance. This is illustrated by the curve schematically shown in FIG. That is, when the true characteristic has a complex characteristic with strong non-linearity like the curve shown by the solid line in FIG. 4, the region where the curve shown by the solid line can be accurately approximated by the low-order model equation is Naturally it will be limited.

  Taking this as an example, the extreme value T of the curve indicated by the solid line is set as the optimum value, and the extreme value is calculated from a quadratic model equation obtained algebraically using three measured values. The application limit of the following model formula will be further described. As shown in FIG. 5, when three points in the vicinity of the fitness value represented by the triangular plot are measured, a curve indicated by a broken line is obtained. The extreme value of the curve indicated by the broken line substantially coincides with the extreme value T of the curve indicated by the solid line. On the other hand, when three points in a wider area than the above-mentioned neighborhood value of the matching value represented by the white circle plot are measured, a quadratic model formula indicated by a dashed line in FIG. 5 is obtained. Then, the extreme value T ′ calculated from the one-dot chain line curve is greatly deviated from the optimum value (extreme value T).

  As described above, in order to obtain an accurate fit value using a low-order model expression, it is desirable to perform measurement in a region near the suitability value in advance. If the measurement points cannot be narrowed down to the vicinity of the matching value in advance, it is necessary to increase the number of measurement points and perform the narrowing based on the measurement result.

  Therefore, in this embodiment, several representative points are extracted from the 120 operating states, and each operational state is obtained by obtaining a prediction formula that defines the relationship between the control parameter and the adaptive value based on the measurement result of the representative points. The fitness value of is estimated. Then, the center point of the control parameter and the upper and lower values thereof are set to the estimated value and the value in the vicinity thereof, and measurement is performed in each of the 120 operating states. In the present embodiment, the reliability is increased by obtaining each of the above model formulas for each operating state using the measurement values obtained in this way. The representative point may be predicted based on a matching value of a similar engine, or the center point may be set by narrowing down in advance using a large number of measurement points.

  Furthermore, in this embodiment, the prediction formula which defined the relationship between the control parameter and the adaptive value is set for each of the three operation regions shown in FIG. This is because the operating region of the on-vehicle engine 10 actually has different properties in each region shown in FIG. 2, and when these three regions are combined to create the above prediction formula, the estimated value is estimated. This is because there is a concern that it cannot be performed with high accuracy.

  For example, misfire is likely to occur in the region of the lowest load and low rotational speed (region near the idle). For this reason, in the same region, if the requirement of the nitrogen oxide emission amount NOx is set strictly in a portion where misfire is difficult to occur, there may be no solution that satisfies them. Therefore, in the vicinity of the idle region, it may be necessary to devise measures such as relaxing the requirements for the nitrogen oxide emission amount NOx and securing a portion where misfire is unlikely to occur. In addition, the engine speed is the same as that in the vicinity of the idle region, and in the racing start region where the engine load is larger, the torque variation is likely to be a problem, and the control parameter that minimizes the torque variation amount TF is an appropriate value. Is desirable. Further, in the normal region, which is a region other than the idle vicinity region and the racing start region, there are often sufficient sets of solutions that satisfy the requirements of the fuel consumption amount BSFC, the nitrogen oxide emission amount NOx, and the torque fluctuation amount TF. Therefore, in this region, it is desirable to set the required condition so as to supplement the condition of the nitrogen oxide emission amount NOx relaxed in the region near the idle.

  From this situation, in the present embodiment, different requirements are set for each of the operation areas, the normal area, the idle vicinity area, and the racing start area. In the present embodiment, the prediction formula for each of the above regions that satisfies these different requirements is defined by the following formula.

<Idle neighborhood area>
_{tang = b 11 × ne + b} 12 × kl + b 13 × ne × kl + b 14
_{Ainj = b 21 × ne + b} 22 × kl + b 23 × ne × kl + b 24 × kl × kl
+ B _{25
vvt = b 31 × ne + b} 32 × kl + b 33 × ne × kl + b 34 × kl × kl
+ B _{35
egr = b 41 × ne + b} 42 × kl + b 43 × ne × kl + b 44 × kl × kl
+ B _{45
aop = b 51 × ne + b} 52 × kl + b 53 × ne × kl + b 54 × kl × kl
+ B ₅₅
<Racing start area>
tang = c ₁₁ × ne + c ₁₂ × kl + c ₁₃ × ne × kl + c _{14
Ainj = c 21 × ne + c} 22 × kl + c 23 × ne × kl + c 24 × ne × ne
+ C ₂₅ × kl × kl + c _{26
vvt = c 31 × ne + c} 32 × kl + c 33 × ne × kl + c 34 × kl × kl
+ C _{35
egr = c 41 × ne + c} 42 × kl + c 43 × ne × kl + c 44 × kl × kl
+ C ₄₅
aop = c ₅₁ × ne + c ₅₂ × kl + c ₅₃ × ne × kl + c ₅₄ × kl × kl
+ C ₅₅
<Regular area>
tang = d ₁₁ × ne + d ₁₂ × kl + d ₁₃ × ne × kl + d ₁₄ × ne × ne
+ D ₁₅ × kl × kl + d ₁₆
Ainj = d ₂₁ × ne + d ₂₂ × kl + d ₂₃ × ne × kl + d ₂₄ × ne × ne
+ D ₂₅ × kl × kl + d _{26
vvt = d 31 × ne + d} 32 × kl + d 33 × ne × kl + d 34 × ne × ne
+ D ₃₅ × kl × kl + _{d36
egr = d 41 × ne + d} 42 × kl + d 43 × ne × kl + d 44 × ne × ne
+ D ₄₅ × kl × kl + d ₄₆
aop = d ₅₁ × ne + d ₅₂ × kl + d ₅₃ × ne × kl + d ₅₄ × ne × ne
_{+ D 55 × kl × kl +} d 56
And the said representative point is set to 13 driving | running states illustrated in FIG. Thereby, when calculating each prediction formula in the above three areas, it is possible to use the fitness values of each of the five points included in those areas. Incidentally, the calculation of each prediction formula using these five matching values is performed as follows, for example.

  First, standard deviations std (ne), std (kl) and average avvte (ne), avvte (kl) of five representative points (ne, kl) are calculated, and the engine speed NE and engine load at these representative points are calculated. KL is defined by the following formula.

NE = {ne-avvte (ne)} / std (ne)
KL = {kl-avvte (kl)} / std (kl)
Then, by setting the average of the appropriate values of the control parameters as avvte (tang), avvte (aop),..., For example, the throttle opening tang can be approximated as the following equation.

tang = d ₁₁ × NE + d ₁₂ × KL + d ₁₃ × NE × KL + d ₁₄ × NE × NE
+ D ₁₅ × KL × KL + avvte (tang)
_{= D 11 × NE + d 12} × KL + d 13 × NE × KL + d 14 × NE × NE
+ D ₁₅ × KL × KL + d ₁₆
In addition, for measurement points near the boundaries of multiple areas, instead of directly using the estimated values obtained based on the prediction formulas for each area, a process in which the estimated values do not differ greatly on both sides of the boundary between the areas. Use the value marked with.

  Such processing can be performed, for example, in the following manner. First, modeling by fuzzy inference is performed, and a membership function that can smoothly connect the vicinity of the boundary of each region is defined in advance. And the value of the membership function corresponding to the estimated value obtained based on each said prediction formula is multiplied. Thereby, the value between the estimated values obtained from each prediction formula can be set as the center point.

Next, calculation of each model formula in each of the 120 driving states, and how to calculate the optimum value of each control parameter in the vehicle-mounted engine 10 alone will be described.
FIG. 7 is a block diagram showing the configuration of a measurement system that calculates such model formulas and control parameters and measures engine characteristic values for the calculation.

  7 includes a dynamometer 31 connected to the output shaft 24 of the in-vehicle engine 10, a dynamometer operating panel 32 for operating the dynamometer 31, and a dynamometer for controlling the dynamometer 31 to predetermined conditions. An automatic measuring device 33 for sending a command to the operation panel 32 is provided.

  The dynamometer 31 is for performing various tests by absorbing the torque generated by the output shaft 24 of the in-vehicle engine 10 so that the in-vehicle engine 10 is mounted on the vehicle in a pseudo state. 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.

  Further, the measurement system shown in FIG. 7 is replaced with the test ECU 30 ′ for controlling the in-vehicle engine 10 instead of the ECU 30 mounted on an actual vehicle, and the test ECU 30 ′ and the automatic measurement device 33. A panel checker 34 that mediates exchange of data between them is provided. The automatic measurement device 33 acquires the measurement information of the in-vehicle engine 10 held in the test ECU 30 ′ via the panel checker 34. The automatic measuring device 33 supplies data corresponding to the accelerator operation amount to the test ECU 30 ′ via the panel checker 34 based on the operating state of the in-vehicle engine 10 monitored by the measurement information.

  That is, when the in-vehicle engine 10 is actually mounted on the vehicle, the driving state is controlled based on the measurement information input to the ECU 30 from the various sensors. On the other hand, when creating a state in which the vehicle is artificially mounted using the dynamometer 31, the automatic measuring device 33 generates the data such as the accelerator operation amount reflecting the driver's intention in a pseudo manner, By supplying to the test ECU 30 ′ via the panel checker 34, the in-vehicle engine 10 is controlled to a desired operation state.

The test ECU 30 ′ includes a calculation map of an engine of a model similar to the in-vehicle engine 10 as control information for the in-vehicle engine 10.
Control commands from the automatic measuring device 33 to the in-vehicle engine 10 and the dynamometer 31 are set based largely on a condition file in the automatic measuring device 33. In the condition file, basically, the control parameters are written for each operation state (engine speed ne and torque) of the in-vehicle engine 10 desired to be measured. At the time of calculating the model formula, the in-vehicle engine 10 is fixedly controlled for each driving state, and the output of the in-vehicle engine 10 during the fixed control is measured by the measuring device 35. Each condition set in the condition file is set by the condition setting tool 53.

  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 in order to control the driving state of the in-vehicle engine 10 in each driving state set in the condition file. When the in-vehicle engine 10 is controlled to the operation state set through this condition file, the automatic measuring device 33 sets a manual flag in the memory or register in the test ECU 30 ′ via the panel checker 34. This manual flag is a flag for prohibiting the control of the in-vehicle engine 10 by the control map. When the in-vehicle engine 10 enters the operation state set through the condition file, the automatic measurement device 33 sets this flag and controls the control parameter of the in-vehicle engine 10 to a value set in the condition file.

  Thus, under the engine operating conditions set in the condition file, the control parameter is fixedly controlled at a predetermined control value, and the fuel consumption BSFC, nitrogen oxide emission NOx in the exhaust, torque fluctuation TF, etc. Each engine characteristic value is measured by the measuring device 35. Specifically, the measuring instrument 35 is a fuel consumption meter that measures the amount of fuel supplied to the in-vehicle engine 10, an analyzer that analyzes the nitrogen oxide concentration in the exhaust gas flowing through the exhaust passage 17 of the in-vehicle engine 10, the in-vehicle engine 10, and the dynamo. A torque meter installed between the meters 31 and a torque variation meter for calculating the value of the torque meter are provided. The automatic measuring device 33 calculates the fuel consumption BSFC based on the measurement value of the fuel consumption meter, the nitrogen oxide emission amount NOx based on the measurement value of the analyzer, and the torque fluctuation amount TF based on the measurement value of the torque fluctuation meter. We are looking for in processing.

  Further, the measurement system includes a server 40 for holding the measurement data for each condition file, an analysis tool 50 for analyzing the measurement data held in the server 40 together with information on each condition file, and the analysis tool 50. A display 51 for displaying the analysis result and a database 52 for storing and holding a part of the analysis result are provided. The measurement system is provided with an operation unit 60 for operating the analysis tool 50, the condition setting tool 53, and the like.

Next, the above-described model formula using the above measurement system and the optimum value calculation procedure will be described. These calculations are performed through the procedure shown in FIG.
That is, when calculating these model formulas and optimum values, first, at step 110, the engine characteristic values at the representative points including the 13 points are measured. This is done in the following procedure.
(A) In the condition setting tool 53, a condition file is set for each representative point.
(B) A value serving as the center point of the control parameter at each representative point is input from the outside via the operation unit 60.
(C) In the condition setting tool 53, for each of the representative points, the input value is set as a center point, and the value of the control parameter used for measurement is set based on the orthogonal table of the experimental design method. The value of the set control parameter is entered in the condition file.
(D) When a condition file is set for all 13 representative points, this condition file is transferred to the automatic measuring device 33.
(E) The automatic measuring device 33 resets the manual flag set in the test ECU 30 ′. In this state, when a predetermined command is sent to the dynamometer 31 and the test ECU 30 ′, the rotational speed of the in-vehicle engine 10 is controlled to match the engine rotational speed set in the specific condition file. Next, control is performed so that the load of the in-vehicle engine 10 is set in the same condition file.
(F) Then, based on the measurement data supplied from the test ECU 30 ′ via the panel checker 34, if it is determined that the operating state of the in-vehicle engine 10 matches the setting in the condition file, the automatic measuring device 33 The manual flag is set in the test ECU 30 ′ via the panel checker 34. At the same time, each control parameter of the in-vehicle engine 10 is fixed to one of 29 types set in the condition file.
(G) Various characteristic values of the in-vehicle engine 10 are measured in this state. When the measurement over a predetermined period is completed, the control parameter is fixedly controlled to another value set in the condition file, and the measurement is performed again.
(H) Thus, when the measurement of the 29 points set in one condition file is completed, the measurement data is automatically registered in the server 40. Then, the next condition file is selected, the manual flag in the test ECU 30 ′ is reset, and the operation state of the in-vehicle engine 10 is controlled to the operation state set in the newly selected condition file. .

  When the measurement of the representative points consisting of 13 operating states is completed by such a series of procedures, the process proceeds to step 120. In step 120, each model formula is calculated for each representative point by the analysis tool 50. That is, first, the analysis tool 50 fetches measurement data from the server 40 and a corresponding condition file from the automatic measurement device 33, respectively. Then, the model formula is calculated based on various measurement conditions such as the type of control parameter at the time of measurement written in the condition file and the measurement data. Then, when the model formula is calculated for each representative point, the process proceeds to step 130.

  In step 130, the analysis tool 50 calculates the optimum value that satisfies the required condition for each representative point from the model formula for each representative point. This required condition is input to the analysis tool 50 from the outside via the operation unit 60 in advance.

  For example, with respect to the representative points belonging to the above-mentioned regular area (see FIG. 6), the required condition sets an upper limit for the nitrogen oxide emission amount NOx and the torque fluctuation amount TF, and the consumed fuel amount is minimum within this range. Is set to be Incidentally, a region where the nitrogen oxide emission amount NOx is less than or equal to the upper limit value and the torque fluctuation amount TF is between the minimum value and the allowable upper limit value from the calculated model formulas is an appropriate region. Therefore, here, a value that minimizes the fuel consumption amount BSFC in this compatible region is calculated as the optimum value. In the racing start area (see FIG. 6), the optimum value is calculated from the graph of the fuel consumption amount BSFC at the time of the minimum torque fluctuation for the reason described above.

  When the optimum value at each representative point is calculated in this way, the measured value, the optimum value, and the model formula used for calculating the optimum value are graphed for each representative point and displayed on the display 51. As a result, the robustness can be checked. That is, the optimum value calculated by numerical analysis by the analysis tool 50 may be a value with poor robustness. Therefore, robustness is checked by graphing and displaying the calculated optimum value, measured value, and the model formula.

  When it is determined that the optimum value calculated for each representative point satisfies the robustness in this way, in step 140, the analysis tool 50 determines each of the idle neighborhood area, the racing start area, and the regular area based on these optimum values. The prediction formula is calculated for each region.

  When the prediction formula is calculated, in step 150, the analysis tool 50 estimates the fitness values of all 120 map points. In this estimation, the value calculated from each prediction formula is not directly used for the vicinity of the boundary between the three regions, and the calculated value is gradually changed based on a membership function or the like as described above. The value obtained is used.

  When the fitness value is estimated at each of the 120 map points in the above aspect, in step 160, the same measurement as in step 110 is performed with these as center points. In other words, after the condition setting tool 53 sets a condition file for each map point, each estimated fitness value is written as a center point in each condition file. Based on this condition file, the automatic measuring device 33 measures each 29 points in which the center point and its upper and lower values are set as control parameter values. Each measurement result is automatically registered in the server 40.

  When the measurement of all 120 driving states is completed in such a series of procedures, the process proceeds to step 170 in FIG. In step 170, each model equation is calculated for each of the 120 operating states in the same manner as in step 200.

  In the subsequent step 180, as in step 130, the optimum value of each control parameter in the 120 operating states is calculated based on the model formula calculated in the previous step 170. The calculated model formulas and the optimum values of the control parameters for each operating state are automatically registered in the server 40.

  The optimum value of each control parameter for each operating state calculated through the above series of procedures is calculated considering only the requirements for each operating state, and is the final evaluation of engine characteristics. It does not always meet the total engine characteristic requirements in the driving mode. Therefore, in the present embodiment, a final control parameter conforming value that satisfies the total engine characteristic requirement in the travel mode is determined by the procedure of Step 300 to Step 800 described above.

  First, the simulation system 70 for determining such a suitable value will be described with reference to FIG. In this embodiment, the simulation system 70 has a configuration corresponding to the control parameter adaptation device, the model formula storage unit, the setting unit, the evaluation unit, the update unit, the re-execution unit, and the determination unit.

  The simulation system 70 is configured as a computer system that executes various processes related to the determination of the adaptive value of each control parameter. The simulation system 70 displays a central processing unit (CPU) that executes arithmetic processing, a storage device that stores data and programs being processed, an input device 71 for inputting conformance conditions, processing results, and the like. Display device and the like. Furthermore, the simulation system 70 is provided with a database 72 in which data of various vehicles, data of travel patterns (changes in vehicle speed, etc.) in various travel modes, and the like are stored in advance. The simulation system 70 receives measurement data calculated by the measurement system and registered in the server 40, that is, data of each model formula and data of optimum values of the control parameters in each operation state. ing.

  FIG. 9 is a block diagram showing the flow of processing related to determination of the fitness value in the simulation system 70. The details of the process related to the determination of the fitness value will be described below with reference to FIG.

  When the processing is started, the CPU of the simulation system 70 receives, from the server 40, the data of each model formula and the data of the optimum value of each control parameter calculated by the measurement system and the simulation system 70. (Step 300 in FIG. 3).

  Subsequently, the vehicle type on which the vehicle-mounted engine 10 to be matched and the running mode for matching are set through the input device 71. In addition, the total engine characteristic requirement in the selected travel mode, for example, the average fuel consumption in the travel mode, the allowable upper limit value of the total emission amount of nitrogen oxides, the allowable limit of torque fluctuation, and the like are also set.

At this time, the CPU fetches the specification data of the vehicle in the selected vehicle type and the data of the selected travel mode from the database 72 into its storage device (step 400 in FIG. 3). For example, the following data is included in the specification data of the vehicle captured here.
・ Weight specifications Vehicle weight, center of gravity height, wheelbase, etc. ・ Transmission specifications Gear ratio, loss characteristics, inertia values, shift patterns, etc. ・ Torque converter specifications, torque converter characteristics, inertia values, etc. ・ Differential specifications Loss characteristics Inertia values, etc., engine characteristics specifications Inertia values, acceleration control time constants, WOT characteristics, etc., tire specifications, tire radius, coefficient of friction with road surface, drag torque during braking, etc.

  On the other hand, the travel mode data includes information on the transition of the vehicle speed at each time of the travel mode. Based on the specification data of these vehicles and the data of the travel mode, the operating state of the in-vehicle engine 10 in the travel mode in which the data is registered can be obtained.

  When the vehicle specification data and the travel mode data are captured, the CPU obtains the operating state of the in-vehicle engine 10 at each time of the travel mode based on the data. That is, here, transition patterns of the engine rotational speed ne and the engine load kl during operation of the in-vehicle engine 10 in the travel mode are obtained.

  Subsequently, the CPU sets each calculation map that is used for evaluation of the validity of matching (step 500 in FIG. 3). In the first stage, the optimum value of each control parameter in each operation state calculated by the measurement system and registered in the server 40 is set as the map value of the calculation map as it is.

  When the calculation map is set, the CPU simulates the behavior of the in-vehicle engine 10 in the travel mode based on the set calculation map, and calculates the total engine characteristic in the travel mode. The calculation of the total engine characteristic here is performed as follows.

  That is, first, based on the operation state of the vehicle-mounted engine 10 at each time in the obtained travel mode, the value of each control parameter at each time is obtained from the set calculation map. Then, the obtained engine parameter values are calculated by substituting the calculated values of the respective control parameters into the above model formula. Based on the engine characteristic value of each period calculated in this way, the total engine characteristic in the driving mode is obtained. For example, the average fuel consumption in the driving mode is obtained by integrating the fuel consumption BSFC at each time determined based on the model formula over the entire driving mode and dividing it by the time from the start to the end of the driving mode. Is required. Further, the total emission amount of nitrogen oxides in the running mode can be obtained by time integration of the NOx emission amount NOx at each time obtained based on the above model formula.

  When the total engine characteristics in the driving mode are thus obtained, the CPU evaluates whether or not the obtained total engine characteristics satisfy the total engine characteristic requirement set at the start of the above processing ( Step 600 of FIG. That is, here, the validity of the map value of the set operation map is evaluated.

  If there is at least one item that has not reached the total engine characteristic request, the CPU updates the map value in the calculation map of each control parameter (step 700 in FIG. 3). That is, the CPU reviews the set values of the control parameters in each operation state so that the unachieved item is achieved.

In this embodiment, the update of the map value of such a calculation map is performed using a comprehensive evaluation function that comprehensively evaluates each engine characteristic value in each operating state. This comprehensive evaluation function is defined by the following equation. The comprehensive evaluation function is individually set for each of the 120 operating states.
Overall evaluation function F (BSFC, NOx, TF) = f ₁ (BSFC) + f ₂ (NOx) + f ₃ (TF)
Here, f ₁ (BSFC), f ₂ (NOx), and f ₃ (TF) are engine characteristic values for each operating state, that is, target fuel consumption amount BSFC, nitrogen oxide emission amount NOx, and torque fluctuation amount TF. It is an evaluation function that indicates the degree of deviation from the value, and is defined by the following equations.
・ Fuel consumption evaluation function f ₁ (BSFC) = k _bsfc × | BSFC−tBSFC |
Nitrogen oxide emission evaluation function f ₂ (NOx) = k _nox × | NOx−tNOx |
・ Torque fluctuation evaluation function f ₃ (TF) = 0 (TF ≦ bTF)
f ₃ (TF) = k _tf × | TF−tTF | (TF> bTF)
Here, tBSFC, tNOx, and tTF indicate target values of the engine characteristic values, respectively. As the target values tBSFC, tNOx, and tTF for each engine characteristic value, for example, an average target value obtained by dividing the total target value for each engine characteristic in the travel mode by the total output can be used.

  However, as such a target value, it is only necessary to set a predetermined value that allows the engine characteristic value to be evaluated to be an undesirable side as the difference between the target value and each engine characteristic value increases. As for the consumed fuel amount BSFC, the nitrogen oxide emission amount NOx, and the torque fluctuation amount TF, all of these target values are “0” because they are engine characteristic values that become undesirably as the value increases. Even if set to “”, the calculation map can be updated without any problem.

  By the way, in the case of an engine characteristic value that becomes less desirable as the value becomes smaller, a sufficiently large target value, for example, a maximum value of the assumed engine characteristic value, etc. may be set as the target value. . Further, in the case of an engine characteristic value that becomes more undesirable as it deviates from a certain value, if the certain value is set as a target value, the same calculation map can be updated.

_Further , k _bsfc , k _nox , and k _tf in the above equation indicate weighting factors of the respective engine characteristic values. This weighting coefficient is a coefficient for enabling the amount of deviation from the target value of different engine characteristic values of each unit system to be quantitatively compared in the same dimension.

Further, bTF indicates a reference torque fluctuation amount that is an upper limit value of the torque fluctuation amount TF whose influence can be ignored. The weighting coefficient k _tf of the evaluation function f ₃ (TF) of the torque fluctuation amount TF is defined as a step function with respect to the reference torque fluctuation amount bTF. That is, when the torque fluctuation amount TF is equal to or less than the reference torque fluctuation amount bTF, the weight coefficient k _tf takes the value “0”, and when it exceeds the reference torque fluctuation amount bTF, the weight coefficient k _tf takes a predetermined constant. This is because the influence of the torque fluctuation amount TF is excluded from consideration in the evaluation as long as it is equal to or less than the reference torque fluctuation amount bTF.

The initial values of the respective weighting factors (k _bsfc , k _nox , k _tf ) are set when the optimum values of the respective operation states calculated by the measurement system are set as the map values of the calculation map. Values that are combinations of control parameters such that the value of the comprehensive evaluation function F () is substantially minimized are set in advance.

  After the evaluation, the CPU adds a predetermined value to the weighting coefficient only for the engine characteristic value related to the item that has not been achieved in the evaluation. And CPU updates the map value of each calculation map using each updated weighting coefficient. At this time, the CPU calculates a combination of control parameters that minimize the value of the comprehensive evaluation function F () based on the updated weighting coefficient in each of the 120 driving states, and calculates them in each of the calculation maps. Set as a map value for the operating state.

10A and 10B show the evaluation functions f ₁ (BSFC), f ₂ (NOx), f ₃ (TF), and the comprehensive evaluation function before and after updating the weighting coefficient based on the result of the evaluation. A relationship between F () and the VVT advance amount vvt, which is one of the control parameters to be updated, is shown. Here, for easy understanding, it is assumed that each engine characteristic value for each operating state is determined only by the VVT advance amount vvt.

As shown in FIG. 10A, before the weighting coefficient is updated, the VVT advance amount vvt at which the value of the comprehensive evaluation function F () is minimum takes the value “A”.
Here, as a result of the evaluation, it is assumed that only the total engine characteristic request relating to the fuel consumption BSFC has not been achieved. At this time, the weighting factors k _nox and k _tf of the evaluation coefficients related to the nitrogen oxide emission amount NOx and the torque fluctuation amount TF satisfying the total engine characteristic requirement are maintained as they are, and the fuel consumption amount BSFC is maintained. Only the weight coefficient k _{bsfc of the} evaluation function related to is added to a predetermined value.

When the weighting factor is updated in this way, as shown in FIG. _10B , the evaluation function f ₁ (BSFC) of the fuel consumption amount BSFC at each VVT advance amount vvt according to the increasing rate of the weighting factor k _bsfc. ) Is increased, and the relationship between each VVT advance amount vvt and the value of the comprehensive evaluation function F () changes. As a result, the value of the VVT advance amount vvt at which the value of the comprehensive evaluation function F () is minimum is advantageous for the fuel consumption amount BSFC for which the total engine characteristic request has not been achieved, that is, the fuel consumption amount BSFC is It changes to a value “B” that can be further reduced.

  Such updating of the map value is performed by performing each of the 120 operation states of each calculation map. As a result, the map value of the calculation map is updated to a value that makes it easier to achieve the total engine characteristic request for items that have not been achieved in the evaluation.

  After updating the map value, the CPU updates the calculation map according to the updated map value, and re-executes the calculation of the total engine characteristic and the evaluation. The processing relating to the update of the map value of the calculation map, the calculation of the total engine characteristic, and the evaluation thereof is repeatedly executed until the total engine characteristic request item is satisfied.

  When it is evaluated that all the items of the total engine characteristic request are satisfied, the CPU determines the map value of the calculation map set at that time as an appropriate value. Then, the calculation map in which the fitness values are stored is output as a file.

In the present embodiment described above, the following effects can be obtained.
(1) In the present embodiment, a calculation that satisfies the total engine characteristic requirement in the driving mode is performed by the simulation system 70 based on the model formula representing the correlation between each control parameter and each engine characteristic value in each driving state. You can adapt the map. Therefore, it is possible to easily adapt the control parameters of the in-vehicle engine 10 in consideration of the vehicle traveling state without repeating trial and error on the engine bench.

(2) Since the update procedure of the map value of the calculation map according to the evaluation result in the simulation system 70 has been established, it is possible to easily automate the processing related to the update.
(3) In the present embodiment, since the above model formula is obtained based on the measurement result of the engine characteristic value in the in-vehicle engine 10, a model formula with high accuracy and reliability can be obtained, and the reliability of the conforming value Can be improved.

  (4) By measuring the engine characteristic values related to the calculation of the model formula based on the experimental design method, while maintaining high accuracy and reliability of the model formula, the man-hours and time required for the measurement are greatly reduced. Can do.

The present embodiment described above may be modified as follows.
-The weight coefficient of the engine characteristic that has not been required may be made variable in accordance with the degree of the failure. For example, the weighting factor may be changed greatly when the requirement is not significantly satisfied, and the weighting factor may be reduced when the requirement can be satisfied in a short time. By doing in this way, the adaptation process can be further accelerated and the adaptation accuracy can be further improved.

  In the above-described embodiment, the update is performed by increasing the weight coefficient of the engine characteristic that has not yet been achieved. However, it is also possible to decrease the weight coefficient of the engine characteristic that has been achieved or both at the same time. You can do the same thing.

  The evaluation function for each engine characteristic value may be a function other than that shown in the above embodiment. In short, if the function takes a value representing the degree of deviation of the engine characteristic value from the target value, it can be adopted as an evaluation function.

-It is good also as a structure which adapts for arbitrary control parameters other than the control parameter illustrated in the said embodiment.
-Arbitrary engine characteristics other than the engine characteristics exemplified in the above embodiment, for example, cooling characteristics, acceleration characteristics, travel characteristics, and the like of the in-vehicle engine 10 may be set as required conditions related to the adaptation of the calculation map.

  -The structure of the vehicle-mounted engine used as the object of a calculation map is not restricted to what was illustrated to the said embodiment.

The schematic diagram which shows the structure of the vehicle-mounted engine used as the application object of the control parameter by one Embodiment of this invention. The figure which shows an example of the calculation map of the control parameter applied to the same vehicle-mounted engine. The flowchart which shows the process sequence which concerns on the adjustment of a calculation map. The figure which shows the setting mode of the control parameter at the time of engine characteristic measurement. The figure which shows the example of an approximation with the low-order model formula of a complicated function. The figure which shows the example of a setting of the driving | operation area | region of a vehicle-mounted engine, and a representative point. The block diagram which shows the structure of a measurement system. The flowchart which shows the process sequence which concerns on the model type | formula and optimal value of each driving | running state. The block diagram which shows together the structure of a simulation system, and the flow of the process. (A) The figure which shows each the relationship between each evaluation function and control parameter after update of a weighting coefficient, and (B) after update.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Car-mounted 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 (VVT), 21 ... throttle valve, 22 ... EGR passage, 23 ... EGR valve, 24 ... output shaft, 25 ... rotational speed sensor, 26 ... water temperature sensor, 30 ... electronic control unit (ECU), 30 '... ECU for testing 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 ... Operating unit, 70 ... Simulation system (control parameter adapting device, model expression storage means, setting hand , Evaluation means, updating means, re-executing means, determination means), 71 ... database.

Claims (7)

A method of adapting control parameters in each driving state so that each of a plurality of engine characteristics of an in-vehicle engine satisfies a total engine characteristic requirement in a predetermined driving mode,
A step of measuring an engine characteristic value by changing the control parameter in each operation state, and calculating a model expression representing a correlation between the control parameter and the engine characteristic value in each operation state based on the measurement value When,
Setting the control parameters for each operating state based on the calculated model formula;
Based on the set control parameter, the engine characteristic value at each time when the vehicle-mounted engine is operated in the travel mode is calculated based on the model formula, and the total engine characteristic is calculated based on the calculation result. Assessing whether requirements are met; and
Updating the control parameter setting mode based on the model formula so that the degree of deviation from the target value of the engine characteristic value that does not satisfy the total engine characteristic requirement is reduced according to the result of the evaluation When,
And repeatedly executing the setting of the control parameter, the updating of the setting mode and the evaluation until the total engine characteristic request is satisfied for all of the plurality of engine characteristics in the evaluation. A control parameter adaptation method for an in-vehicle engine that acquires an adaptation value of the control parameter. Setting of the control parameters based on the model equation, A _1, A _2, ..., a A _n and each engine characteristic values, the deviation degree of f _m the (A _m) from the target value for any of the engine characteristic values A _m Is an evaluation function that represents
F (A ₁ , A ₂ ,..., A _n ) = f ₁ (A ₁ ) + f ₂ (A ₂ ) +... + F _n (A _n )
Is performed by calculating the control parameter of each operating state where the value of the comprehensive evaluation function represented by
The update of the control parameter is performed according to the result of the evaluation so that the degree of reflection of the engine characteristic value that does not satisfy the requirements in the evaluation with respect to the value of the comprehensive evaluation function becomes larger. The method for adapting a control parameter of an in-vehicle engine according to claim 1, which is performed by correcting the evaluation function. The T _m and the target value for any of the engine characteristic values A _m, when the k _m and the weight coefficient of the engine characteristic values A _m, the evaluation function f _m of the engine characteristic values A _m (A _m) is
_{f m (A m) = k} m × | A m -T m |
Represented by
The evaluation function is modified so that the ratio of the weight coefficient of the engine characteristic value that does not satisfy the total engine characteristic request to the weight coefficient of the engine characteristic value that satisfies the total engine characteristic request increases. The control parameter adaptation method for an in-vehicle engine according to claim 2, which is performed by correcting a weight coefficient of the characteristic value. The control parameter adaptation method according to any one of claims 1 to 3, wherein the measurement of the engine characteristic value is performed by determining measurement points based on an experimental design method. In the in-vehicle engine control parameter adaptation device that adapts the control parameter in each driving state so that each of the plurality of engine characteristics of the in-vehicle engine satisfies the total engine characteristic requirement in a predetermined driving mode,
Model equation storage means for registering and storing a model equation representing a correlation between each engine characteristic value and the control parameter in each operation state;
Setting means for setting the control parameter in each operating state based on the model formula;
Based on the control parameter, the engine characteristic value at each time when the vehicle-mounted engine is operated in the travel mode is calculated based on the model formula, and the total based on the calculated engine characteristic value at each time An evaluation means for evaluating whether each of the engine characteristic requirements is satisfied,
When at least one of the total engine characteristic requests is not satisfied in the evaluation by the evaluation unit, the setting unit is configured to reduce the degree of deviation from the target value of the engine characteristic value that does not satisfy the engine characteristic request. Updating means for updating a setting mode of the control parameter based on the model formula;
A re-execution unit configured to re-execute the setting of the control parameter of the setting unit and the evaluation of the evaluation unit when the setting mode is updated by the updating unit;
A determination unit that determines the control parameter at that time as an appropriate value when the total engine characteristic request is satisfied for all of the plurality of engine characteristics in the evaluation of the evaluation unit;
In-vehicle engine control parameter adaptation device comprising: The setting means, A _1, A _2, ..., when the A _n and each engine characteristic values, and f _m and (A _m) an evaluation function that represents the deviation degree from the target value for any of the engine characteristic values A _m ,
F (A ₁ , A ₂ ,..., A _n ) = f ₁ (A ₁ ) + f ₂ (A ₂ ) +... + F _n (A _n )
The control parameter for each operating state is set based on the model formula so that the value of the overall evaluation function represented by
The update means modifies the evaluation function of each engine characteristic value so that the degree of reflection of the engine characteristic value that is not satisfied with the total engine characteristic requirement with respect to the value of the total evaluation function is larger, The control parameter adaptation device for an in-vehicle engine according to claim 5, wherein the setting mode is updated. The T _m and the target value for any of the engine characteristic values A _m, when the k _m and the weight coefficient of the engine characteristic values A _m, the evaluation function f _m of the engine characteristic values A _m (A _m) is
_{f m (A m) = k} m × | A m -T m |
Represented by
The modification of the evaluation function is to increase the ratio of the weight coefficient of the engine characteristic value not satisfying the total engine characteristic request to the weight coefficient of the engine characteristic value satisfying the total engine characteristic request. The on-vehicle engine control parameter adaptation device according to claim 6, wherein the control parameter adaptation device is performed through correction of a weighting factor of the vehicle engine.

Images