Method for simulation-based analysis of a motor vehicle

Abstract

The invention relates to a method for simulation-based analysis and/or optimization of a motor vehicle, preferably having the following working steps: simulating (SIOI) a driving operation of the motor vehicle (I) on the basis of a model (M) with at least one manipulated variable for acquiring values of at least one simulated variable which is suitable for characterizing an overall vehicle behaviour, in particular a driving capability, of the motor vehicle (I), wherein the model has at least one partial model, in particular a torque model, and wherein the at least one partial model is based on a function and preferably characterizes the operation of at least one component, in particular of an internal combustion engine of the motor vehicle (I); and—outputting (S I03) the values of the at least one simulated variable.

Claims

1. A method for simulation-based analysis or optimization of a motor vehicle, comprising: simulating a driving operation of the motor vehicle on the basis of a function-based model with at least one manipulated variable for acquiring values of at least one simulated variable suitable for characterizing an overall vehicle behavior of the motor vehicle, wherein the function-based model comprises at least one partial model, wherein the at least one partial model is based on a first function that characterizes an operation of a drive apparatus of the motor vehicle, wherein the at least one partial model is a torque model of the drive apparatus of the motor vehicle, and wherein the at least one partial model comprises at least one of the following sub-models: a full load model based on a full-load function, whereby at low engine speed and medium engine speed the full-load function is approximated by torque, while at maximum power the full-load function is approximated by a curve to power; a torque gradient model based on a torque gradient function which has a linear portion and a cubic portion, and wherein a function parameter of the torque gradient model indicates a weighting of the linear portion and the cubic portion; and a partial-load model based on a partial-load function which is calculated based on the full-load function and a pedal characteristic function, wherein the pedal characteristic function indicates a correlation between a torque variable and a pedal variable or a throttle valve position variable; and outputting the values of the at least one simulated variable.

2. The method according to claim 1, further comprising: determining one or more driving mode parameters defined in relation to one or more values of the at least one simulated variable and the at least one manipulated variable, the one or more driving mode parameters characterizing at least one driving mode, and wherein the values of the at least one simulated variable are output in conjunction with a respective driving mode parameter of the one or more driving mode parameters.

3. The method according to claim 1, wherein the at least one partial model is based on a second function comprising at least one function parameter, and wherein modifying the at least one function parameter allows the simulated driving operation of the motor vehicle to be modifiable.

4. The method according to claim 2, wherein the respective driving mode parameter is defined by at least one predetermined condition in relation to at least one of the at least one manipulated variable and the at least one simulated variable.

5. The method according to claim 1, wherein simulating the driving operation is performed for various points of an experimental design.

6. The method according to claim 2, further comprising: determining at least one evaluation parameter value of at least one evaluation parameter, wherein the at least one evaluation parameter indicates the overall vehicle behavior of the motor vehicle based on an assignment rule dependent on one or more of the at least one simulated variable and the respective driving mode parameter; and outputting the at least one evaluation parameter value.

7. The method according to claim 6, further comprising: preparing a first technical specification with respect to the driving operation of the motor vehicle, wherein the first technical specification corresponds to a setpoint range for the respective driving mode parameter or an evaluation parameter of the at least one evaluation parameter corresponding to target values for design criteria for the overall vehicle behavior of the motor vehicle.

8. The method according to claim 7, further comprising: optimizing the at least one partial model based on the first function in relation to the setpoint range for the respective driving mode parameter or the evaluation parameter corresponding to the target values for the design criteria for the overall vehicle behavior of the motor vehicle.

9. The method according to claim 6, further comprising: adapting the one or more values of the at least one simulated variable for the respective driving mode parameter or the at least one evaluation parameter to a predefined setpoint range; and modifying at least one function parameter of a second function of the at least one partial model used for the simulated driving operation if the adapted one or more values of the at least one simulated variable or the at least one evaluation parameter lie outside of the predefined setpoint range, wherein the method repeats the simulating of the driving operation; and/or outputting a value of the at least one function parameter of the second function of the at least one partial model used for the simulated driving operation if the adapted one or more values of the at least one simulated variable or the at least one evaluation parameter lie within the predefined setpoint range.

10. The method according to claim 9, wherein the modification of the at least one function parameter of the second function occurs based on an optimization algorithm, and wherein the at least one function parameter of the second function used for the simulated driving operation is treated in the optimization algorithm as the at least one manipulated variable of the drive apparatus or the motor vehicle.

11. The method according to claim 3, further comprising: generating an experimental design which comprises points of variation in regard to the at least one function parameter of the second function used for the simulated driving operation based on an optimization algorithm, wherein the simulated driving operation occurs based on the experimental design.

12. The method according to claim 9, further comprising: defining a second technical specification for the drive apparatus or the motor vehicle based on the second function used for the simulated driving operation or the value of the at least one function parameter.

13. The method according to claim 12, wherein the at least one partial model characterizes the drive apparatus of the motor vehicle, and the method further comprises: designing or modifying a design, a control, or a regulation of the drive apparatus or the motor vehicle based on the second function used for the simulated driving operation or the value of the at least one function parameter.

14. The method according to claim 1, wherein the torque model comprises at least the full load model, and wherein the full-load function specifies a full-load characteristic based on three subfunctions: a first full-load function subfunction at the low engine speed; a second full-load function subfunction at the medium engine speed; and a third full-load function subfunction at the maximum power.

15. The method according to claim 1, wherein the pedal characteristic function comprises a first function parameter and a second function parameter, wherein the first and second function parameters are speed-dependent, and wherein the first function parameter indicates a first factor and the second function parameter indicates an offset.

16. The method according to claim 1, wherein the at least one partial model comprises at least one function parameter as a manipulated variable, by the manipulating of which the simulated driving operation of the motor vehicle can be changed.

17. The method according to claim 6, wherein one or more of the respective driving mode parameter and the at least one evaluation parameter is determined as a function of a vehicle parameter.

18. The method according to claim 1, wherein the model comprises a vehicle model as a further partial model, and wherein the further partial model is configured to at least partly characterize a driving characteristic of the motor vehicle.

19. The method according to claim 1, further comprising: providing vehicle parameters with respect to the motor vehicle, wherein the driving operation of the motor vehicle is simulated by a vehicle model.

20. A non-transitory computer readable medium containing instructions which when executed by a computer cause the computer to perform a method comprising: simulating a driving operation of the motor vehicle on the basis of a function-based model with at least one manipulated variable for acquiring values of at least one simulated variable suitable for characterizing an overall vehicle behavior of the motor vehicle, wherein the function-based model comprises at least one partial model, wherein the at least one partial model is based on a first function that characterizes an operation of at least one component of the motor vehicle, and wherein the at least one partial model comprises a partial-load model based on a partial-load function which is calculated based on a full-load function and a pedal characteristic function, wherein the pedal characteristic function indicates a correlation between a torque variable and a pedal variable or a throttle valve position variable; and outputting the values of the at least one simulated variable.

21. A system for simulation-based analysis or optimization of a motor vehicle, comprising: a first module configured to simulate a driving operation of the motor vehicle on the basis of a function-based model with at least one manipulated variable for acquiring values of at least one simulated variable suitable for characterizing an overall vehicle behavior of the motor vehicle, wherein the function-based model comprises at least one partial model, wherein the at least one partial model is based on a first function that characterizes an operation of at least one component of the motor vehicle, and wherein the at least one partial model comprises a partial-load model based on a partial-load function which is calculated based on a full-load function and a pedal characteristic function, wherein the pedal characteristic function indicates a correlation between a torque variable and a pedal variable or a throttle valve position variable; a second module configured to output the values of the at least one simulated variable, wherein the first module and the second module are connected by a first data interface; and hardware configured to perform at least one of the first module and the second module.

22. The system according to claim 21, wherein the second module is configured to determine one or more driving mode parameters defined in relation to one or more values of the at least one simulated variable and the at least one manipulated variable, the one or more driving mode parameters characterizing at least one driving mode, and wherein the one or more values of the at least one simulated variable are output in conjunction with a respective driving mode parameter of the one or more driving mode parameters.

23. The system according to claim 21, wherein the at least one partial model is based on a second function comprising at least one function parameter, and wherein modifying the at least one function parameter allows the simulated driving operation of the motor vehicle to be modifiable.

24. The system according to claim 21, wherein the first data interface is configured to furnish: vehicle parameters and values of at least one of the at least one manipulated variable and the at least one simulated variable from the first module to the second module; and values of a function parameter and points of variation from the second module to the first module.

25. The system according to claim 22, wherein the second module is configured to determine at least one evaluation parameter, wherein the at least one evaluation parameter indicates the overall vehicle behavior of the motor vehicle based on an assignment rule dependent on one or more of at least one output simulated variable and the respective driving mode parameter, and wherein the second module is further configured to output the at least one evaluation parameter.

26. The system according to claim 25, wherein the second module is configured to prepare a first technical specification with respect to the driving operation of the motor vehicle, wherein the first technical specification corresponds to a setpoint range for the respective driving mode parameter or an evaluation parameter of the at least one evaluation parameter corresponding to target values for design criteria for the overall vehicle behavior of the motor vehicle.

27. The system according to claim 26, wherein the second module is configured to optimize the at least one partial model based on the first function in relation to setpoint range for the respective driving mode parameter or the evaluation parameter corresponding to the target values for the design criteria for the overall vehicle behavior of the motor vehicle.

28. The system according to claim 25, wherein the second module is further configured to: adapt the one or more values of the at least one simulated variable output for the respective driving mode parameter to a predefined setpoint range; modify at least one function parameter of a second function of the at least one partial model used for the simulated driving operation if the adapted one or more values of the at least one output simulated variable lie outside of the predefined setpoint range, wherein the first module is configured to repeat the simulating of the driving operation; and/or output a value of the at least one function parameter if the adapted one or more values of the at least one output simulated variable lie within the predefined setpoint range.

29. The system according to claim 25, further comprising a third module that is connected to the second module via a second data interface and to the first module via a third data interface, wherein the third module is configured to: adapt the at least one evaluation parameter to a predefined setpoint range; modify at least one function parameter of a second function of the at least one partial model used for the simulated driving operation if the at least one adapted evaluation parameter lies outside of the predefined setpoint range, wherein the first module is configured to repeat the simulating of the driving operation; and/or output a value of the at least one function parameter if the at least one adapted evaluation parameter lies within the predefined setpoint range.

30. The system according to claim 29, wherein the second data interface is configured to furnish values of the at least one evaluation parameter from the second module to the third module and the third data interface is configured to furnish the values of the at least one function parameter and points of variation from the third module to the first module.

31. The system according to claim 29, wherein the second module or the third module is configured to generate an experimental design which comprises the points of variation in regard to the at least one function parameter of the second function used for the simulation based on an optimization algorithm, wherein the simulation occurs based on the experimental design.

Description

(1) Shown at least partly schematically therein are:

(2) FIG. 1 a first exemplary embodiment of the inventive system;

(3) FIG. 2 a second exemplary embodiment of the inventive system;

(4) FIGS. 3 to 6 an inventive subdividing of a full-load characteristic of an internal combustion engine into three ranges;

(5) FIG. 7 a full-load characteristic with applied approximation function;

(6) FIG. 8 a torque gradient approximation function;

(7) FIGS. 9 and 10 a pedal characteristic approximation function of a partial load model;

(8) FIG. 11 a full-load characteristic with applied partial-load approximation function; and

(9) FIG. 12 a flow chart of an exemplary embodiment of the method according to the invention.

(10) The following will describe a first exemplary embodiment of the inventive system 10 for the simulation-based analysis and/or optimization of a motor vehicle and an associated inventive method 100 on the basis of FIGS. 1 and 12.

(11) The system 10 according to the invention thereby preferably comprises three modules 11, 12, 13, each connected by data interfaces for data transmission. In particular, as indicated by the arrow in FIG. 1, data is transferred from the first module 11 to the second module 12 via a first data interface 14, from the second module 12 to the third module 13 via a second data interface 15, and in turn from the third module 13 to the first module 11 via a third data interface 16.

(12) A model M with which the driving operation of a motor vehicle 1 can be simulated is stored in the first module 11. As indicated by the arrow, this model M can be read in or written into the module 11 from outside via an interface. Preferably, however, the model M can also be read out again from the module 11.

(13) In order to implement a working step of simulation S101 for the first time, a predetermined driving cycle is preferably stored in the first module 11 which represents a sequence of driving modes for the motor vehicle 1. This driving cycle is preferably developed on the basis of experiences made by test engineers and comprises load points for the calibration of a motor vehicle 1, in the present example embodiment, the drive or respectively internal combustion engine of the motor vehicle 1, that experience has shown to be necessary.

(14) The model M makes use of a partial model to simulate the operation of the internal combustion engine 1. This partial model is function-based; i.e. it is specified by a function which comprises function parameters, in particular coefficients, and variables, in particular manipulated variables such as e.g. the engine speed and accelerator pedal position. The function thereby reflects an assignment rule which continuously assigns a value of one or more simulated variables to a set of manipulated variable values.

(15) Preferably, further partial models of the model M specify the function of further components of the vehicle. If these further components are not likewise to be analyzed or optimized by the inventive method 100, then the further partial models can preferably be map-based; i.e. the assignment rule inherent to these partial models is not stored as a function but as a characteristic map which discretely assigns a value of one or more simulated variables to a set of manipulated variable values in each simulation time step.

(16) In particular, the model M can comprise a vehicle model as the further partial model which is configured to simulate a driving characteristic of the motor vehicle 1. The vehicle parameters which enter into such a vehicle model are in particular the weight and/or the size or respectively the center of mass of the vehicle.

(17) A vehicle model is prepared in a known manner in order to be able to simulate a driving operation. The first module 12 thereby preferably simulates a two-mass or multi-mass oscillator in order to reflect the mass of the motor vehicle 1, the rigidity of the power train and the response characteristic of the tires. Further preferably, the damping can also be simulated on the basis of the vehicle model. The damping values depend in particular on the operating state of the vehicle.

(18) Moreover, such a vehicle model can comprise further sub-models such as, for example, a tire model, a wheel suspension model (spring/damper), a geometrical chassis suspension model, a traction resistance model, a steering model, a clutch model, a transmission model and/or an elasticity model (multi-mass system/multi-mass oscillator).

(19) The working step of simulation S101 creates simulated variable values of the model M by executing the predetermined driving cycle on the vehicle 1 characterized by the model M. The initial values of the function parameters or respectively coefficients of the function used for the simulation are thereby preferably selected on the basis of test engineer experiences. Further preferably, these can also, as stated further below with respect to later iterations, be initially prespecified by the third module 13, preferably in the form of points of variation of an experimental design. This type of simulation of the driving operation of the motor vehicle 1 can in particular be realized with the applicant's AVL VSM™ system.

(20) At least one of the simulated variables is thereby suitable, in particular in conjunction with other simulated variables, for characterizing an overall vehicle behavior of the vehicle 1, respectively assessing the overall vehicle behavior based on said variable. In particular, the overall vehicle behavior encompasses at least the driving capability. In this case, the overall vehicle behavior, in particular the driving capability, serves as a criterion for the design of the vehicle 1.

(21) The simulated variables are preferably passed on to the second module 12. The second module 12 is thereby preferably capable of checking the values of the simulated variables S102 for the presence of a given condition. Such a condition is in particular a set of the values of multiple simulated variables and/or the sequence of values of one or more variables. If such a condition is fulfilled, the second module 12 then determines a driving mode and establishes a driving mode parameter thereto.

(22) Alternatively, the driving mode parameter is defined on one or more values of at least one simulated variable or at least one manipulated variable, although does not however represent a separate value but is instead substantially an allocation of values of at least one simulated variable characterizing the overall vehicle behavior of the vehicle to the values of at least one simulated variable and/or at least one manipulated variable characterizing the driving mode.

(23) Additionally to simulated variables, manipulated variables of the model M can also be used to define the driving mode parameter. An example of this is the accelerator pedal and/or throttle valve position, from the value of which a driving mode can be concluded.

(24) The driving mode parameter is thereby preferably a numeric value or a set of numeric values or also defined by a term which is assigned to the value or values.

(25) To determine the driving mode parameter, a database can in particular be provided in the second module 12, on the basis of which the current driving mode can be determined based on the driving mode parameter by adapting values of the simulated and/or manipulated variables.

(26) Data is thereby preferably exchanged between the first module 11 and the second module 12 via the first interface 14, which can be in the form of software and/or hardware.

(27) In the present exemplary embodiments, the values of the at least one simulated variable which characterize the overall vehicle behavior in conjunction with the respectively given driving mode parameter are output. Preferably, the values are thereby directly output to a third module 13 which in particular serves in the applying of an optimization algorithm on the results determined in the simulation.

(28) Further preferably, the values are output to an evaluation algorithm in the second module 12, with which the simulated variables determined for the motor vehicle 1 which express the overall vehicle behavior can be evaluated, preferably objectified. To that end, in particular utilized is an assignment rule between at least one simulated variable, by means of which the overall vehicle behavior can be characterized, and the driving mode parameter to the evaluation parameter into which enters the evaluation of one or more motor vehicles by human test drivers, particularly as regards reference vehicles. Establishing such an assignment rule by which to sort an evaluation parameter as a function of a simulated variable, by means of which the overall vehicle behavior can be characterized, will be clarified below by means of an exemplary embodiment relative to the driving mode. This relates to a tip-in in second gear, thus an acceleration process with increasing throttle opening.

(29) One exemplary embodiment of establishing an assignment rule for the assigning of an evaluation parameter to a respective driving mode parameter or driving mode respectively will be clarified below on the basis of the driving mode of a so-called tip-in in second gear, thus an acceleration process with increasing throttle opening.

(30) In an actual driving operation with a test subject as vehicle occupant, the throttle valve position, the engine speed and the longitudinal acceleration is first measured as a function of time for a tip-in driving mode. Parallel thereto, the subjective perceptions of the test subject are recorded, for example by the test subject inputting their subjective perception as an assessment via a user interface. Preferably a ten-part scale from outstanding=10 to extremely poor=1 can serve as the assessment criteria.

(31) In real time or after recording a dataset, the engine speed n and the longitudinal acceleration is evaluated. Preferably, a Fast Fourier Transformation (FFT) of the engine speed n and the longitudinal acceleration is thereby computed. Moreover, a maximum value of bucking oscillations in the frequency range between 2 and 8 hertz as well as the frequency at which the maximum value occurs is preferably computed according to the following equation:

(32) $a\left(s\right)={\int }_{-\infty }^{\infty }{e}^{-\mathrm{ist}}a\left(t\right)\mathrm{dt}$

(33) st thereby represents the imaginary part and a(t) the chronological course of the acceleration.

(34) From that, a correlation is made according to the following equation between the test subject's subjective perception, the FFT and the maximum bucking oscillation values:

(35) $\mathrm{Dr}=\frac{c1}{\sqrt{c2*a_{\mathrm{osc}}^{c3}}}$

(36) c1, c2 and c3 are thereby parameters, a.sub.osc the maximum value of the bucking oscillation in the 2 to 8 hertz range, and Dr the calculated evaluation parameters, in the present case a so-called driving capability index for the driving capability criterion. The c1, c2 and c3 parameters can preferably be found automatically in a self-learning system. Preferably, iteration loops are used to that end in which the parameters are modified until there is minimum subjective deviation between the legitimate value Dr and the subjective perception of the test subject Dr. This occurs pursuant to the following equations:
c1.sub.i+1=c1.sub.i+p.sub.i,
c2.sub.i+1=c2.sub.i+q.sub.i,
c3.sub.i+1=c3.sub.i+r.sub.i,

(37) Here, the p.sub.i, q.sub.i, and r.sub.i expressions represent variation increments. The c1, c2 and c3 variation ensues until the difference between the calculated evaluation parameter Dr and the subjective evaluation parameter Dr.sub.subj is less than a predefined limit.

(38) After complete system training, the subjective evaluation in the vehicle can be fully simulated from the amplitudes a.sub.osc of the bucking oscillation. The identified parameters c1, c2, c3 reflect the subjective assessment.

(39) The depicted exemplary embodiment for establishing the assignment rule for the evaluation parameter is only one of numerous possibilities for creating the assignment rule. The iteration can thus also be realized with other known mathematical or statistical methods.

(40) Alternatively, the assignment rule can also be a comparison of a simulated variable to a setpoint range. The setpoint range in this case corresponds to target values for a criterion. For example, a certain fuel consumption over a setpoint range could thereby be given as the target value. The fuel consumption, which is simulated by means of model M, can then be compared to the setpoint range and it thus be determined whether there is an excess or even a shortfall and a change is necessary to the model M or the function parameters contained therein.

(41) In this case, the optimization algorithm is preferably provided in the second module 12 which then creates a new experimental design or set of new variation points of the function parameter respectively to the setpoint range as target value directly on the basis of the simulated variable, the simulated consumption in the present case. The evaluation parameter is in this case the consumption. The simulation is then run again S101 in the first module 11 with the new function parameters.

(42) Alternatively, such a relationship of simulated fuel consumption as simulated variable and the associated setpoint range as target values could also be transformed into an assessment expressed by a numerical value, in particular a grade or a phrase (“too low,” “too high,” “ok”). This evaluation parameter, which is preferably computed by the second module 12, is then preferably output to the third module 13 via the second interface 15. The optimization algorithm stored in this case in the third module 13 then computes function parameter variation points for optimizing consumption, in particular on the basis of an experimental design.

(43) In order to be able to determine the evaluation parameter on the basis of the assignment rule, it is preferably further provided for the second module 12 to provide vehicle parameters with regard to the motor vehicle 1 simulated by the first module 11. These are preferably the mass and the engine characteristic, in particular maximum power, maximum torque, engine speed at maximum power, engine speed at maximum torque and maximum engine speed of the simulated motor vehicle 1. Further preferably, this data is transmitted from the first module 11 to the second module 12 via the first data interface 14.

(44) As shown in the lower diagram in FIG. 1, the evaluation parameter can not only contain a single assessment but can also consist of a plurality of assessments. In FIG. 1, an overall assessment for acceleration at full load consists for example of a maximum expected momentum, a 90% torque threshold, a 90% torque range, a broad torque level, a vehicle turning response, an expected acceleration and a reference acceleration. The second module 12 can thereby be given a plurality of criteria on the basis of which the assessment or respectively the determining of the evaluation parameter S104 for the overall vehicle behavior of the motor vehicle 1 is to be made. Examples of such criteria include e.g. driving capability, agility, driving comfort, emissions, efficiency, NVH behavior, vehicle turning response, sense of safety and function of the driver assistance systems. A particularly accurate assessment of the overall vehicle behavior can thereby be achieved when these criteria are evaluated in a holistic context.

(45) A value of the evaluation parameter determined by means of the assignment rule is then output S105 from the second module 12 to the third module 13 via a second data interface 15 as an alternative to at least one simulated variable in conjunction with the driving mode parameter. Alternatively or additionally, the evaluation parameter can also be output via a user interface.

(46) An optimization algorithm for improving the assessment of the overall vehicle behavior is then preferably run In the third module 13. Function parameters or respectively coefficients of the function of the at least one partial model of model M are thereby input into this evaluation algorithm as variables. These variables are varied on the basis of the optimization algorithm in order to achieve an optimization S107 of the evaluation parameter or the one or more valuation criteria respectively. The criteria of the evaluation can thereby preferably be weighted differently.

(47) Further boundary conditions further preferably enter into the optimization algorithm. This can for example be properties which were not taken into account during the evaluation in the second module 12. Such boundary conditions can for example be a desired torque or a desired performance or even boundary conditions which while not characterizing the overall vehicle behavior of the motor vehicle 1 are nonetheless for example relevant to safety or prescribed by law.

(48) Preferably, prior to modifying the function parameters of the function of the at least one partial model used for simulation, it is determined whether the overall vehicle behavior is already achieving a desired evaluation S106. To this end, particularly values of the at least one simulated variable, by which the overall vehicle behavior can be characterized, are compared to a setpoint range for the respective associated driving mode parameters, in particular with target values for a configuration of the motor vehicle 1. Alternatively or additionally, the evaluation parameters determined by the second module 12 can also be compared to a setpoint range.

(49) The evaluation algorithm is in this case only implemented if a setpoint range has not yet been reached. If, in contrast, the setpoint range has been reached, the last value used of the at least one function parameter of the function of the partial model used for simulation is output S109 as is described in the following.

(50) In particular, the function parameters of the partial model functions used for simulation in the optimization algorithm are treated as manipulated variables, in particular as sole manipulated variables, of the component of the motor vehicle 1 or the function of the partial model of said component respectively.

(51) The function parameters are thereby provided to the third module 13 from the first module 11 via the third data interface 16 or in particular defined as a variable by a user when setting up the optimization algorithm prior to executing the inventive method 100.

(52) If the setpoint range of the evaluation has not yet been reached, the third module 13 then preferably provides an experimental design based on the optimization algorithm S108 which comprises further points of variation in a variation range spanned by the at least one function parameter of the function of the partial model used for the simulating. The driving operation of the motor vehicle 1 is then re-run on the basis of the modified partial model in the manner of an iterative optimization; i.e. with modified function parameters or coefficients, S101′.

(53) Such an experimental design is in particular established on the basis of statistical methods and corresponds to a statistical experimental design (design of experiment). The variation points of such an experimental design are for example the “design points” depicted in FIG. 1.

(54) If, on the other hand, it is determined during the comparison with respect to the assessment S106 that the assessment already corresponds to a desired setpoint range in terms of overall vehicle behavior, then the value of the at least one function parameter or respectively coefficient of the function of the partial model is output S109.

(55) Preferably, the value or values can be output by means of a user interface; further preferably, the value or values are applied in the function used for the simulation.

(56) The function obtained in this way indicates that operational mode of the component of the motor vehicle 1 which it must have in order to achieve a specific assessment of the overall vehicle behavior of the motor vehicle 1.

(57) Under certain circumstances, interactions exist with other components of the motor vehicle 1. Known strategies of multi-variable optimization can preferably be employed in order to take such interaction into account.

(58) On the basis of the function or functions obtained, a technical specification can now be drawn up S110a for the at least one component of the motor vehicle 1 or for the entire motor vehicle 1. In particular, the designs and/or control or regulation of the component of the motor vehicle 1 can be adapted S110b on the basis of the functions obtained. The respective component is thereby preferably designed, configured and controlled so as to reflect an actual operation of the function or respectively function parameters or coefficients as output.

(59) In one particularly preferential exemplary embodiment, the at least one component of the motor vehicle 1 or the entire motor vehicle 1 is not only modified but also fully optimized to the optimized partial model or its function respectively.

(60) For example, a torque model as depicted in FIGS. 3 to 11 can be optimized in respect of a wide variety of criteria relative to the driving operation of a motor vehicle 1. In order to achieve the desired target criteria values, such as for example an expected acceleration and reference acceleration or a certain broad torque level or vehicle turn response, it may for example prove necessary to electrically charge an internal combustion engine of a vehicle; i.e. run the turbocharger additionally or exclusively to an electric motor or even provide an additional electric drive in order to obtain the desired performance parameters of the criteria. This enables a completely new developmental approach in which design engineers can already be drawing conclusions in early stages of development as to how individual components of a motor vehicle 1 need to be designed with defined vehicle parameters in order to later meet criteria related to the driving operation of the motor vehicle 1.

(61) For example, a first technical specification can be prepared by Technical Sales which establishes target values, in particular a setpoint range relative to a criterion, for example the above-cited driving dynamic parameter or at least one evaluation parameter representing the driving dynamic. On the basis of this technical specification, a second technical specification for a component can be determined by means of the inventive method which enables at least one rough design of the component of a motor vehicle 1.

(62) FIG. 2 shows a second exemplary embodiment of the inventive system 10.

(63) In contrast to the first exemplary embodiment, the system 10 depicted in FIG. 2 comprises only one first module 11 and one second module 12. Accordingly, there is also only one first data interface 14 with which data can be exchanged between the first module 11 and the second module 12. In contrast to the first exemplary embodiment, no evaluation parameter is computed as already discussed in conjunction with the first embodiment. Instead, the values of the at least one simulated variable for the respective driving mode parameter, by means of which the overall vehicle behavior can be characterized, are directly compared to a setpoint range for said driving mode parameter in the comparison working step S106.

(64) If the desired target value setpoint range is not reached, the function parameter(s) of the functions used for simulation are modified using any given boundary conditions without an evaluation based on a preferably objective assignment rule being made.

(65) FIGS. 3 to 11 will be drawn on in explaining the developing of a function-based engine model or torque model able to be used as a partial model for illustrating the operation of an internal combustion engine in the inventive method 100.

(66) Particularly depicted with respect to FIGS. 3 to 6 is the developing of a function-based sub-model M1 which illustrates the torque characteristic of a supercharged internal combustion engine 1 in stationary and transient operating range.

(67) Further sub-models of the function-based partial model for the engine are a partial load model M3, see FIG. 9, as well as a torque gradient model M2 and preferably also a suction torque model as well as any further sub-models as necessary.

(68) Characteristic map-based models are customarily used in the prior art to illustrate the functioning of an engine. Thus, a characteristic map is used to illustrate the torque characteristic in the stationary and transient operation of a supercharged internal combustion engine with which the engine torque as currently applied to the crankshaft in each time step of a simulation is determined as a function of load or respectively throttle valve position and engine speed. To illustrate the transient torque build-up, for example after an erratic change in load or respectively accelerator pedal position, characteristic maps are again used for the suction torque and for the torque gradients based on the load pressure build-up of the turbocharger as a function of the initial parameters of accelerator pedal position, engine speed and load at the time of the load change. Spontaneously reached torque of a supercharged internal combustion engine, the so-called quick torque availability, is modeled with a suction torque map whereas the substantially slower torque build-up starting from suction torque up until reaching stationary torque is illustrated with an operating point-dependent torque gradient map.

(69) An engine model consisting of these characteristic map-based sub-models in principle enables simulating with sufficient accuracy transient assessment-relevant driving modes such as for example full load acceleration, low-end torque, turning response, positive load change (tip-in), accelerating and pulling power of a motor vehicle 1.

(70) In order to make such a characteristic map-based engine model of an automated optimization available by means of an optimization algorithm, the invention replaces individual sub-models of the characteristic map-based engine model or even all of the sub-models with function-based sub-models.

(71) The number of variables to be varied for an efficient optimization process, thus the function parameter or respectively the coefficients of the sub-models, can thereby be reduced substantially and the function parameter or coefficients of the individual sub-models can be modified independently of each other, particularly in the context of variable optimization.

(72) Using sub-models particularly enables the torque characteristic of the internal combustion engine to be illustrated in stationary and transient operation by way of mathematical functions having few function parameters or coefficients respectively.

(73) Full Load Model M1

(74) In order to develop a full load model M1 as shown in FIG. 3, the full-load characteristic of an internal combustion engine is observed in three separate segments within a speed range of the internal combustion engine. The three speed segments are identified in FIG. 3 using three different cross-hatchings. The segment up to approximately 2000 rpm can be termed the full-load segment at low engine speed, the segment up to approximately 5200 rpm the full-load segment at medium engine speed, and the segment up to approximately 7000 rpm, the full load segment at maximum power.

(75) This classification was selected in this case so as to correspond as closely as possible to the usual behavior of modern supercharged internal combustion engines, in particular gasoline engines. Preferably also possible, however, are other classifications which are better suited to illustrating the operation of other engines.

(76) The individual segments of the full-load characteristic are hereby specified by way of the engine speed-dependent torque or power curve.

(77) A detailed description of the individual segments with the respectively specified function parameters and mathematical formulas will be provided on the basis of FIGS. 4 to 6. General symbols are thereby used in the formulas as follows:

(78) M=torque in Nm

(79) P . . . =power output in kW

(80) n . . . =rpm in min.sup.−1

(81) Full-Load Segment at Low Engine Speed

(82) The torque in this segment is specified by the following function VF1:

(83) $M_1\left(n\right)=M_{1000}+\frac{n-1000}{1000}.Math.k_{M1}$

(84) The power output then accordingly results as follows:

(85) $P_1\left(n\right)=M_1\left(n\right)*\frac{\pi .Math.n}{30000}$

(86) The individual function parameters are thereby defined as follows:

(87) M.sub.1000=torque at 1000 min.sup.−1

(88) k.sub.M1=torque increase in this segment

(89) As depicted in FIG. 4, the position and slope of the approximation function changes to the full-load characteristic subject to the values of this M.sub.1000 and k.sub.M1 function parameter.

(90) Full-Load Segment at Medium Engine Speed

(91) The torque in this segment can be specified by the following function VF2:

(92) $M_2\left(n\right)=M_{\mathrm{ma}x}+\frac{n-n_M}{1000}.Math.k_{M2}$

(93) The individual function parameters in this segment are thereby defined as follows:

(94) M.sub.max=maximum torque

(95) n.sub.M=rpm at maximum torque

(96) k.sub.M2=torque increase in this segment

(97) Since with positive values of the function parameter k.sub.M2 and at rpm greater than n.sub.M, torques greater than the defined maximum torque are mathematically possible with this formula, the result is limited to M.sub.max. The resulting function for the torque within this range is as follows:
M.sub.2(n)=min(M.sub.2(n),M.sub.max)

(98) Based on this function, the power output is calculated, analogously to the low rpm segment, as follows:

(99) $P_2\left(n\right)=M_2\left(n\right).Math.\frac{\pi .Math.n}{30000}.$

(100) The approximation function to the rpm characteristic in the second segment is depicted in FIG. 5 by function parameters M.sub.max, n.sub.M and k.sub.M2.

(101) Full-Load Segment at Maximum Power

(102) Unlike with the low and medium rpm segment, it is not the torque characteristic being specified in the maximum power segment by a function with function parameters but rather the power curve.

(103) The power curve can thereby be specified in this segment by the following function VF3:

(104) $P_3\left(n\right)=P_{\mathrm{ma}x}+\frac{n-n_P}{1000}.Math.k_{P1}-{\left(\frac{n-n_P}{1000}\right)}^2.Math.k_{P2}$

(105) The function parameters in this segment are thereby defined as follows:

(106) P.sub.max=maximum power

(107) n.sub.P=rpm at maximum power

(108) k.sub.P1=slope to power curve

(109) k.sub.P2=curvature to power curve

(110) As in the medium rpm segment, given positive values of the function parameter k.sub.P1 and rpm greater than n.sub.P, power greater than the defined maximum power of the internal combustion engine are also mathematically possible here based on the function. The result is thus also limited with respect to this function to P.sub.max. The overall function for illustrating the power curve in this segment is therefore as follows:
P.sub.3(n)=min(P.sub.3(n),P.sub.max)

(111) Inversely to the low and medium rpm segment, the torque characteristic of the internal combustion engine 1 is now calculated from the power curve on the basis of the following formula:

(112) $M_3\left(n\right)=P_3\left(n\right).Math.\frac{30000}{\pi .Math.n}\text{:}$

(113) The approximation function to the power curve with the function parameters of P.sub.max, n.sub.P, k.sub.P1 and k.sub.P2 is depicted in the third segment in FIG. 6.

(114) In order to obtain the complete function-based torque model, the model components of the individual segments or respectively the approximation functions underlying the model components are joined together into an overall approximation function for illustrating the entire full load curve.

(115) This preferably ensues by identifying intersection points between approximation functions of the individual segments or by the so-called minimum principle:
M(n)=min(M.sub.1(n),M.sub.2(n),M.sub.3(n))

(116) This principle can be fully understood on the basis of the FIG. 7 graphical representation in which the individual approximation functions VF1, VF2, VF3 of the individual segments are depicted.

(117) Torque Gradient Model M2

(118) In order to define the transient operational behavior of a supercharged internal combustion engine, information is needed on how quickly a turbocharger can build up a necessary charging pressure for providing a high torque. The dependency is inventively defined as torque increase per unit of time and is a function subject to the respectively given engine speed.

(119) The torque gradient is thus preferably inventively specified as a function of the engine speed by the following functions:

(120) $x=\frac{n-1000}{4000}$$y=x.Math.\left(1-\mathrm{prog}\right)+x^3.Math.\mathrm{prog}$$\mathrm{Gradient}\left(n\right)=y.Math.\left({\mathrm{grad}}_5-{\mathrm{grad}}_1\right)+{\mathrm{grad}}_1$

(121) The torque gradient function is hereby defined by the following function parameters:

(122) grad.sub.1=torque gradient at 1000 min.sup.−1

(123) grad.sub.5=torque gradient at 5000 min.sup.−1

(124) prog=progression factor

(125) x, y=auxiliary variables

(126) The auxiliary variables x and y serve only in simplifying the calculation.

(127) The “prog” function parameter of the torque gradient function can be adjusted in a range of 0 to 1 and influences the ratios in the above functions to which the torque gradient function is composed of a linear component and a cubic component.

(128) An inventive approximation function for the torque gradient is depicted in FIG. 8. The torque gradient value is thereby depicted for each engine speed during a pedal jump from 0 to 100%. The grad.sub.1, grad.sub.5 and prog function parameter are likewise depicted in FIG. 8. With respect to the prog function parameter, it is also evident from the arrows how a change in this function parameter affects the course of the torque gradient function.

(129) Partial Load Model M3

(130) The partial load range of a supercharged internal combustion engine depends in particular on the internal combustion engine's pedal characteristic. This defines the correlation between an accelerator pedal position and a torque demand.

(131) Therefore, a function-based partial load model is preferably determined on the basis of a function-based pedal characteristic. Such a function-based pedal characteristic is a percentage, with the help of which the full-load characteristic can be scaled according to the pedal position. The function-based pedal characteristic is inventively specified by the following functions:

(132) 0$x=\frac{\mathrm{Pedal}}{100}$$\mathrm{sShape}=\left(\left(\frac{3}{8}.Math.{\left(2x-1\right)}^5-\frac{5}{4}.Math.{\left(2x-1\right)}^3+\frac{15}{8}.Math.\left(2x-1\right)\right)+1\right).Math.50$$\mathrm{offset}=\left(1-{\left(2x-1\right)}^2\right).Math.\mathrm{shift}$$\%\mathrm{Moment}=\mathrm{Pedal}.Math.\mathrm{linear}+\mathrm{sShape}.Math.\left(1-\mathrm{linear}\right)+\mathrm{offset}$

(133) The mathematical result according to this function can also take values below 0% or above 100%. The result therefore needs to be limited to a valid range of values.

(134) An actual pedal characteristic moreover exhibits an engine speed dependency. For this reason, the two function parameters of the “linear” and “shift” pedal characteristic function are specified as engine speed-dependent functions. To that end, the values of the function parameters are defined at three different engine speeds and the course of the pedal characteristic function then interpolated with a quadratic polynomial.

(135) The progression of such a pedal characteristic function is thereby depicted in FIG. 9 with the “shift” and “linear” function parameters.

(136) The function-based pedal characteristic model thereby exhibits the following function parameters:

(137) linear=linearity

(138) shift=shift

(139) The arrow in FIG. 9 indicates how the approximation function of the pedal characteristic shifts upon a respective change in “shift” and “linear” function parameter. A direct response to an accelerator pedal position (straight dotted line) will thereby be perceived by a driver as rather sporty; in contrast, a more progressive curve of low torque requirement in the lower accelerator pedal position range and higher torque requirement in the upper accelerator pedal position range will be perceived as rather equable.

(140) FIG. 10 depicts the engine speed dependency of the function parameter of the pedal characteristic function, with the different progressions of the pedal characteristic function being depicted for different engine speeds.

(141) Lastly, FIG. 11 shows an inventive partial load model M3 which was calculated from the inventive function-based full load characteristic model and the inventive function-based pedal characteristic model.

(142) In order to achieve better driving capability, it is preferable in calculating the partial-load approximation function at low speeds not to scale the current full load torque but rather the maximum torque. Otherwise, the breaks in the full load characteristic at approximately 2000 min.sup.−1 and 4000 min.sup.−1 would also be reflected in the partial load characteristics.

(143) Accordingly modified, the depicted sub-models and their model components for full load models M1, torque gradient models M2 and partial load models M3 can also be carried over to non-charged internal combustion engines. Other partial models can be accordingly function-based created for other drive systems, for example electric motors, and for other components of the vehicle, for example the steering or the transmission, so as to be able to be optimized with the inventive method 100.

(144) It should furthermore be noted that the exemplary embodiments as depicted are merely examples which are in no way to be construed as limiting the protective scope of the invention's application and structure. Rather, the foregoing description gives the skilled person a guide in implementing at least one exemplary embodiment, whereby various modifications, particularly with respect to the function and arrangement of the described components, may be made without departing from the invention's scope of protection as yielded by the claims with these equivalent feature combinations.

LIST OF REFERENCE NUMERALS

(145) 1 motor vehicle 10 system 11 first module 12 second module 13 third module 14 first interface 15 second interface 16 third interface M model M1, M2, M3 partial model