JP2006259975A - Method for designing vehicle, including tire - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To decide the cornering the characteristics of tires which can be realized by actual tires as request characteristics in the design of a vehicle, including the tires. <P>SOLUTION: The values of parameters B-E when the lateral force of a tire and the characteristic curve of a self-aligned torque are approximated by "Magic Formula" are applied to a vehicle model, and traveling simulation is carried out for performance evaluation. When predetermined performance is not satisfied by the vehicle model in the performance evaluation, the traveling simulation is carried out, by correcting the values of the parameters B to E for the performance evaluation of a vehicle; and the characteristic curve of the tire to be specified by the values of the corrected parameters B to E is calculated, and tire dynamic element parameters are derived, based on the tire dynamics model configured, by using a plurality of tire dynamics element parameters from the characteristic curve. When the predetermined performance is satisfied by the vehicle model, however, the tire dynamics element parameters, corresponding to the parameters B to E, are decided as being the tire request characteristics. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention provides vehicle specifications (wheel base, height of center of gravity, total weight of vehicle, weight distribution of front and rear wheels of vehicle, etc.) and tires in order to realize desired vehicle performance by performing vehicle running simulation. The present invention relates to a design method of a vehicle including a tire that determines a required characteristic of the vehicle.

Currently, in the initial stage of vehicle development, determination of vehicle specifications that achieve desired vehicle performance is performed by performing a simulation using a computer and determining whether to achieve desired vehicle performance. That is, a vehicle model is generated based on the set vehicle specifications, a tire driving force is applied to the vehicle model, a vehicle running simulation is performed, and the performance evaluation of the vehicle in the vehicle specifications is performed from the simulation result.
In particular, when performing a running simulation of a vehicle, the tire is the only component that transmits the force received from the road surface to the vehicle, and is a component that greatly affects the performance of the vehicle. For this reason, it is necessary to accurately determine the generated force of the tire.
A characteristic curve representing the slip angle dependence of the lateral force and self-aligning torque generated during tire cornering is approximated by the “Magic Formula” represented by the following basic formula (9) as the generated force of the tire. It has been proposed to do.
Y (x) = Dsin [Ctan ⁻¹ {Bx−E (Bx−tan ⁻¹ (Bx))}] (9)
“Magic Formula” is a non-analytic model that uses a nonlinear approximation formula that represents tire characteristics by determining the values of parameters B to E in Formula (9).
In general, a vehicle running simulation is performed using the values of parameters B to E of the “Magic Formula”.

FIG. 17 is a flowchart showing the flow of a vehicle design method using such vehicle performance evaluation.
First, at the initial stage of vehicle development, the desired vehicle specifications (wheelbase, height of center of gravity, weight distribution of the front and rear wheels of the vehicle, etc.) are set and a vehicle model is generated. A value of B to E is set and given to the vehicle model. Next, a travel condition for a vehicle travel simulation is set, and a vehicle travel simulation is performed according to the travel condition. The traveling conditions differ depending on the performance to be evaluated. For example, when the performance evaluation is durability evaluation, profile data such as the traveling speed of the vehicle and the roughness of the road surface is set as the traveling conditions. In the case of the emergency avoidance performance evaluation, data such as vehicle traveling speed and steering angle, actual road surface profile data, and the like are set as traveling conditions.
On the other hand, the vehicle model is, for example, a mechanism analysis model used in mechanism analysis software such as ADAMS (manufactured by MDI, USA), CarSim (manufactured by Mechanical Simulation Corporation), or MatLab (manufactured by The MathWorks, Inc.). An analysis model used in vehicle motion analysis software combined with software for control design can be mentioned.

  Next, performance evaluation data serving as a performance index is extracted based on the result of the running simulation. For example, in the case of durability performance, the maximum value of stress applied to a specific member and the deformation amount of the specific member are used as performance evaluation data. In the case of emergency avoidance performance, the stability based on the time trace of the relationship between the traveling speed of the vehicle and the maximum lateral acceleration or the slip angle of the vehicle and the rate of change thereof is evaluated. If the performance evaluation data does not reach the target value determined in the vehicle planning stage, a part of the vehicle specifications is corrected. For example, the weight distribution of the front and rear wheels of the vehicle is corrected, the characteristics of the components such as the suspension are corrected, and the generated force of the tire such as the tire longitudinal force characteristic curve and the lateral force characteristic diagram is corrected. When correcting the generated force of the tire, the values of the parameters B to E of the “Magic Formula” described above are corrected.

Thus, the correction is repeated until the performance evaluation data reaches the target value. If the target value is achieved, the values of the parameters B to E of “Magic Formula” representing the vehicle specifications, the characteristics of the components and the tire characteristics at that time are determined as required characteristics.
The car manufacturer moves on to detailed design of the component according to this decision. In the case of a tire that cannot be designed by itself, the determined required characteristics of the tire, that is, the values of the parameters B to E of the “Magic Formula” are presented as one of the required items to the tire manufacturer, and the tire delivery is ordered.
The tire manufacturer performs structural design of the tire and material design of the tire that realize the values of the parameters “B” to “E” of the “Magic Formula”, and repeats trial manufacture of the tire so as to meet the presented requirement items.

  However, in the above method, the parameters B to E of “Magic Formula” used as the required characteristics of the tire are parameters of a model specialized for describing the cornering characteristics while ignoring the actual mechanical behavior of the tire. For this reason, the parameters B to E shown in the above equation (9) do not reflect the structural mechanical mechanism that affects the cornering characteristics of the tire. For this reason, even if the values of the parameters B to E are given as required items, there is no connection with the tire dynamic element parameters representing various stiffnesses of the tire that affect the cornering characteristics of the tire, and the structural design and material design of the tire There was a problem that it was not reflected directly. Further, since the parameters B to E of the “Magic Formula” given as requirements are determined without considering the structural mechanical mechanism of the tire, it can be said that it can be realized with an actual tire. Often not.

On the other hand, Patent Document 1 below discloses a tire design method capable of performing an optimum design by evaluating tire performance in combination with a vehicle on which a tire is mounted.
Specifically, a tire is modeled with a finite element based on a design value, and a tire running simulation is performed. By this simulation, tire characteristic data is obtained, and a vehicle running simulation is performed using the tire characteristic data. . In this method, since the tire is modeled finely by a finite element, there are an extremely large number of parts to be corrected, and there are many cases where it is unclear where to correct, and it is necessary to rely on a skilled designer. In addition, it is necessary to repeat tire running simulation and vehicle running simulation many times, and it is difficult to efficiently find design values that realize satisfactory performance.

Further, in “Magic Formula”, as shown in Non-Patent Document 1 below, by changing the scale factor λ, the lateral force and the longitudinal force generated in the tire can be corrected, that is, the tire characteristics are corrected. Can do. That is, the lateral force and the longitudinal force generated in the tire can be corrected independently. For example, in order to greatly increase the stopping distance performance at the initial stage of development, the tire's longitudinal force is set to be large, and the lateral force is set to be lowered to prevent rollover when an emergency failure is avoided, and this is determined as a required characteristic of the tire. can do.
However, since the lateral force and longitudinal force generated in the tire are not forces generated by an independent mechanism from the viewpoint of structural mechanics, but are closely related forces, the above required characteristics of the tire are realized in an actual tire. In many cases, it is impossible.

JP 2002-356106 A “TYRE AND VEHICLE DYNAMICS”, (HANS. B. PACEJKA, ISBH 0-7506-5141-5), pp.172-191

  Therefore, in the present invention, unlike conventional vehicle design methods, approximate expression parameters in an approximate expression such as “Magic Formula” expressed as tire cornering characteristics are combined with tire dynamic element parameters including various tire rigidity. In addition, it provides a design method for vehicles including tires that can realize the values of approximate formula parameters such as “Magic Formula” to be determined, and present tire requirement characteristics that can be efficiently realized. For the purpose.

  The present invention is a vehicle running simulation using information on characteristic curves representing slip ratio dependence of force or torque acting on a tire rotation axis based on shearing force acting on a road surface and information on vehicle specifications. A vehicle design method including tires for designing a vehicle having a desired vehicle performance by performing a model creation step for creating a vehicle model using information on vehicle specifications, and nonlinearly converting the characteristic curve When approximated by an approximate expression, a value of an approximate expression parameter that defines a non-linear approximate expression is assigned to the vehicle model, a travel simulation is performed under a predetermined travel condition, and a vehicle performance evaluation is performed using the result of the travel simulation If the vehicle model does not satisfy the target performance in the performance evaluation step and the performance evaluation, the vehicle specification information is corrected and the corrected vehicle specifications are corrected. A vehicle specification correction step of generating a vehicle model using the information of the vehicle and performing the travel simulation and the performance evaluation of the vehicle, and in the performance evaluation, if the vehicle model does not satisfy a predetermined target performance, the approximate expression The parameter value is corrected, the corrected approximate expression parameter value is assigned to the vehicle model, a travel simulation is performed, the vehicle performance is evaluated using the result of the travel simulation, and the approximate expression parameter value is corrected. A characteristic curve is calculated from a nonlinear approximation formula defined by the equation, and based on a tire dynamic model configured using a plurality of tire dynamic element parameters, the tire dynamic element parameter value defining the characteristic curve is calculated from the characteristic curve. Deriving tire characteristic correction step, and correcting if the vehicle model satisfies the target performance Tire characteristic determination step for determining the tire dynamic element parameter value derived corresponding to the approximate expression parameter value as a tire required characteristic, and the tire characteristic correction step includes the vehicle specification correction step. Provided is a method for designing a vehicle including a tire, which is performed after performing the steps at least once.

The characteristic curve is a curve representing the slip angle dependency of the lateral force and self-aligning torque generated on the tire shaft when a slip angle is given, or the longitudinal force generated on the tire shaft when the rolling speed of the tire is adjusted. Including a curve representing the slip ratio dependency of.
When the characteristic curve is a characteristic curve representing the slip angle dependency of the lateral force generated on the tire shaft and the self-aligning torque, the tire dynamic model calculates the lateral force and calculates the self-aligning torque on the tire. It is preferable that the self-aligning torque be calculated by dividing the lateral force torque component generated by the lateral force acting on the ground contact surface and the longitudinal force torque component generated by the longitudinal force acting on the tire ground contact surface.
In the tire characteristic correction step, when deriving the value of the tire dynamic element parameter, the sum of square residuals of the characteristic curve of the lateral force and the corresponding curve of the lateral force calculated by the tire dynamic model, The weighted coefficient is a value obtained by weighting and adding a square residual sum of a characteristic curve of the self-aligning torque and a corresponding curve of the self-aligning torque calculated by the tire dynamic model using a weighting coefficient. As a composite square using a coefficient obtained from information on variation in values depending on the slip angle of the characteristic curves of the lateral force and the self-aligning torque calculated from the values of the approximate expression parameters The value of the tire dynamic element parameter indicating the characteristic during cornering of the tire is set so that the value of the residual sum is not more than a predetermined value. It is preferable to leave.
Further, in the tire characteristic correction step, when deriving the value of the tire dynamic element parameter from the characteristic curve based on the tire dynamic model, it is given by the torsional deformation of the tire generated by the self-aligning torque. It is preferable to derive the value of the tire dynamic element parameter using the effective slip angle in which the slip angle is corrected.

The tire dynamic element parameters derived in the tire characteristic correction step include, for example, an adhesion friction coefficient between the tire tread member and the road surface, a sliding friction coefficient, and a shape defining coefficient that defines the shape of the contact pressure distribution. At that time, in the tire characteristic correction step, using at least one of the rigidity parameter for tire shear deformation, the rigidity parameter for lateral bending deformation of the tire, and the rigidity parameter for torsional deformation of the tire, which are obtained in advance, It is preferable to derive the adhesion friction coefficient, the sliding friction coefficient, and the shape defining coefficient.

In the present invention, an approximate expression parameter is given to the vehicle model to perform a travel simulation. If the vehicle model does not satisfy the target performance in the performance evaluation at that time, the approximate expression parameter is corrected and the travel simulation is performed again. At this time, a characteristic curve is calculated from a nonlinear approximation formula defined by the corrected approximation formula parameter, and a tire dynamic element parameter is derived from the characteristic curve based on the tire dynamic model. For this reason, the approximate expression parameters in the nonlinear approximate expression expressed as the cornering characteristics of the tire are linked to the tire dynamic element parameters including various stiffnesses of the tire, and can be determined as parameters that can be realized in an actual tire. In addition, since the tire dynamic element parameters are derived each time the approximate expression parameter is corrected, when the vehicle model satisfies a predetermined performance, the tire dynamic element parameters that are easy to understand for the tire manufacturer are also determined. This tire dynamic element parameter can be efficiently presented as a tire required characteristic.
Furthermore, tire characteristics have a great influence on the performance of the vehicle. For this reason, the tire characteristic correcting step for correcting the tire characteristics is performed after at least one vehicle specification correcting step for correcting the vehicle specification information and performing the travel simulation and the vehicle performance evaluation. Then, the portion where the target performance cannot be achieved can be efficiently achieved by the tire characteristics, and the vehicle including the tire can be efficiently designed.

  Hereinafter, a vehicle design method including a tire according to the present invention will be described in detail based on embodiments shown in the accompanying drawings.

FIG. 1 is a configuration diagram of an apparatus 1 for carrying out a vehicle designing method including a tire according to the present invention. The apparatus 1 includes a computer that performs a vehicle design method by executing various programs.
The device 1 obtains desired performance from data on vehicle specifications (wheel base, height of center of gravity, total weight of vehicle, weight distribution of front and rear wheels of the vehicle, etc.) and information on cornering characteristics of tires that are components of the vehicle. Is a device for determining vehicle specifications and tire required characteristics for realizing a vehicle having the following.

  The apparatus 1 includes a CPU 2 that manages and controls the execution of each part of the computer and each program, a memory 4 that stores various conditions and calculation results via the bus 3, and a mouse and a keyboard that inputs various conditions and various information. An input operation system 5 such as an interface 6, an interface 6 for connecting the input operation system 5 to the bus 3, and an output device for displaying processing results of various programs including various conditions and information input screens and simulation results, and printing them out. 7 and a program group 8 having various programs to be described later and realizing the functions of the apparatus 1.

Here, the program group 8 includes a setting program 9 that sets values of parameters “B” to “E” of “Magic Formula” as vehicle specification data and tire cornering characteristics, and vehicle travel that performs vehicle travel simulation using a vehicle model. Given the simulation program 10 and the values of the “Magic Formula” parameters B to E, the tire cornering characteristic curve data is calculated. When the cornering characteristic curve is given, the “Magic Formula” parameter B When a “Magic Formula” data parameter calculation program 11 for calculating the value of E and a characteristic curve of cornering are given, a plurality of tire dynamic element parameters (hereinafter referred to as dynamic element parameters) based on a tire dynamic model described later. ) Or the dynamic element parameters in the tire dynamic model Tire dynamic model program group 12 that calculates cornering characteristic curves of lateral force and self-aligning torque (hereinafter referred to as torque) using a tire dynamic model, and the control and management of these programs are integrated. And an integration / management program 13 that corrects the values of the parameters B to E of the “Magic Formula” according to the simulation result and evaluates the performance of the vehicle from the simulation result.
Here, the cornering characteristic curve is a curve representing the slip angle dependency of lateral force and torque.
The parameters B to E of “Magic Formula” are approximate expression parameters that define a nonlinear approximate expression when the characteristic curves of the lateral force and the self-aligning torque are approximated by the nonlinear approximate expression shown in the above formula (9). It is.

The setting program 9 is a part for setting the data of the vehicle specifications and the values of the parameters B to E of “Magic Formula”. For example, the wheel base, the height of the center of gravity, the total weight of the vehicle, the weight distribution of the front and rear wheels of the vehicle, Data such as suspension characteristics is set via the input operation system 5 or by calling data stored in the memory 4. In addition, the values of the parameters B to E of “Magic Formula” are set by calling data stored in the memory 4.
The vehicle travel simulation program 10 creates a vehicle model according to the data of the vehicle specifications, calls the travel simulation condition given from the operation input system 5 or the travel condition stored in the memory 4, and creates the generated vehicle model. This is a portion for performing the vehicle running simulation by giving the values of the set parameters B to E.

The vehicle model used for the traveling simulation is a mechanism analysis model, for example, a model created by mechanism analysis software ADAMS. Alternatively, an analysis model defined by the motion analysis software CarSim or the control system design software MatLab may be used.
The traveling simulation conditions are the traveling speed of the vehicle, the steering angle, the profile shape of the road surface, and the like, and different traveling simulation conditions are set according to the performance to be evaluated.
The running simulation is performed by mechanism analysis software ADAMS. Further, the calculation of the running simulation may be performed by the motion analysis software CarSim or the control system design software MatLab.

The “Magic Formula” data / parameter calculation program 11 uses the values of the parameters B to E to determine the values of the lateral force and the self-aligning torque according to the above equation (9) when the values are given to the parameters B to E. Calculate and obtain characteristic curves of lateral force and self-aligning torque. The values of parameters B to E are given for several types of loads, and a characteristic curve having a slip angle dependency is obtained for each load. Further, by using the values of parameters B to E given for each load, it is possible to obtain a characteristic curve of the load dependency of the lateral force and the self-aligning torque when the slip angle is 1 degree.
On the other hand, when the characteristic curve depending on the slip angle dependency of the lateral force and the self-aligning torque is given to the “Magic Formula” data parameter calculation program 11, the values of the parameters B to E are obtained from this characteristic curve. The method for obtaining the values of the parameters B to E is not particularly limited. For example, since the equation (9) is nonlinear with respect to the parameters B to E, it is preferably performed according to the Newton-Raphson method.
The tire dynamic model program group 12 will be described in detail later.

  The integration / management program 13 integrates control and management of the above program, and corrects the data of the vehicle specifications and the values of the parameters “B” to “E” of the “Magic Formula” according to the vehicle simulation result. In addition, the vehicle performance is evaluated from the simulation result. If the predetermined evaluation data calculated as performance evaluation does not reach the set target value, the integration / management program 13 uses the vehicle specification data and the values of parameters B to E as the target value for which the evaluation data is set. Repeat until you are satisfied. The correction method is not particularly limited. For example, the data of the vehicle specifications and the values of parameters B to E are changed within a predetermined range. When the evaluation data calculated as the performance evaluation does not reach the set target value, the dynamic element parameters derived from the corrected vehicle specification data, the values of parameters B to E, and the values of parameters BE to be described later Are determined as design specifications that achieve the desired performance. The mechanical element parameter is determined as a required characteristic of the tire. Such required characteristics of the tire are output to the output device 7.

  The tire dynamic model program group 12 represents a tire dynamic model by an analytical expression, and calculates a lateral force and a torque using the set dynamic element parameters, and a predetermined tire dynamic model calculation program 14 to the tire dynamic model calculation program 14. It includes a program for deriving various dynamic element parameters by calculating in sequence, or calculating lateral force and torque in which the force is in a balanced state (equilibrium state) with a tire dynamic model. The various mechanical element parameters derived by the tire dynamic model program group 12 or the calculated lateral force and torque characteristic curves are output to the output device 7.

Examples of the dynamic element parameters calculated based on the tire dynamic model are as follows.
(A) the lateral stiffness K _y0 defined by the lateral shear stiffness of the tire,
(B) sliding friction coefficient between road surface and tire μ _d ,
(C) lateral stiffness coefficient (K _y0 / μ _s ) obtained by dividing lateral stiffness K _y0 by an adhesion friction coefficient μ _s between the road surface and the tire,
(D) the lateral bending coefficient ε of the belt member,
(E) Torsional compliance (1 / _Gmz ) which is the reciprocal of torsional rigidity around the tire central axis of the tire,
(F) Coefficient n that defines the contact pressure distribution on the contact surface during the generation of lateral force,
(G) a coefficient C _q representing the degree of deflection of the ground pressure distribution,
(H) a movement coefficient C _xc indicating the degree of movement of the tire center position in the front-rear direction on the ground contact surface;
(I) Effective ground contact length l _e when lateral force is generated,
(J) The longitudinal rigidity A _x in the contact surface (a parameter that determines the longitudinal force torque component), and the like.

Here, lateral stiffness K _y0, lateral bending coefficient epsilon, G _mz of the torsional compliance (1 / G _mz), the rigid parameters for shear deformation of the tire, respectively, with respect to torsional deformation of the stiffness parameter and the tire with respect to the lateral bending deformation It is a stiffness parameter. Further, the lateral direction in which the lateral force is generated means the axial direction of the tire rotation axis. Therefore, the lateral direction when the tire rolls in a straight line state is the left-right direction with respect to the rolling direction, and the direction coincides. However, when the slip angle is added, the lateral direction is the slip angle with respect to the rolling direction of the tire. I can't tell. The front-rear direction refers to a direction that is parallel to the road surface on which the tire contacts and that is orthogonal to the axial direction of the rotation axis of the tire. The tire central axis (referred to as the axis CL in FIGS. 5A and 5B) is perpendicular to the road surface perpendicular to the rotational axis of the tire and passing through the central plane in the tire width direction. Is the axis.

The tire dynamic model program group 12 has four different types of sequences, and CP (lateral force when the slip angle is 1 degree) and SATP (self-alignment when the slip angle is 1 degree) corresponding to each sequence. CP / SATP parameter calculation program 20 for determining the respective mechanical element parameters from the load dependence curve of torque), and F _y / determining the respective dynamic element parameters from the characteristic curve of the lateral force and torque slip angle dependence. _Mz parameter calculation program 22, CP / SATP data calculation program 24 for obtaining calculation data of CP and SATP in a state of force balance in the tire dynamic model, and lateral force and torque in a state of force balance in the tire dynamic model has a _F y / _{M z} data calculating program 26 to obtain the calculation data, the It is made.
The functions of the CP / SATP parameter calculation program 20, the F _y / M _z parameter calculation program 22, the CP / SATP data calculation program 24, and the F _y / M _z data calculation program 26 will be described later.

The tire dynamic model calculation program 14 is the vehicle specification data set in the setting program 9 and the values of the parameters B to E of the “Magic Formula”, or the vehicle specification data corrected by the integration / management program 13. Then, using the values of parameters “B” to “E” of “Magic Formula”, side force and torque correspondence calculation data (lateral force F _y ′ and torque M _z ′) are calculated based on the tire dynamic model, and the calculated values are processed. As a result, the operation unit returns to the CP / SATP parameter calculation program 20, the F _y / M _z parameter calculation program 22, the CP / SATP data calculation program 24, or the F _y / M _z data calculation program 26.
2, 3, 4 (a) to (c), FIGS. 5 (a) to (d), and FIGS. 6 (a) to (c) are diagrams illustrating a tire dynamic model.

  As shown in FIG. 2, the tire dynamic model includes a sidewall model composed of a plurality of spring elements representing the spring characteristics of the sidewalls on a rigid cylindrical member, and a belt model composed of an elastic ring body connected to these spring elements. And a tread model composed of an elastic element representing a tread model connected to the surface of the elastic ring body.

In the tire dynamic model defined by the tire dynamic element calculation program 14, more specifically, as shown in FIG. 3, various linear parameters configured by combining various spring elements of the tire and lateral bending of the belt member. Dynamic element parameters including non-linear parameters such as coefficient ε and coefficient C _q are set, and slip angle α, lateral force F _y and torque M _z are input, and equations (1) to (6) in FIG. The processed lateral force and torque values (hereinafter referred to as lateral force F _y ′ and torque M _z ′) are calculated. Of course, an error between the input lateral force F _y and torque M _z and the processed lateral force F _y ′ and torque M _z ′ is equal to or less than a predetermined value, that is, substantially coincides (converged tire dynamics). Only when the force is balanced in the model), the values of the lateral force F _y 'and the torque M _z ' are determined as the values of the lateral force and torque of the tire that realize the force balanced state.
The linear parameter means a dynamic element parameter expressed in a linear form in the equations (5) and (6), and the non-linear parameter is expressed in a non-linear form in the equations (5) and (6). It is a dynamic element parameter.

The tire dynamic element calculation program 14 calculates a torsion return angle obtained from the torque M _z and the torsion compliance (1 / G _mz ) input based on the formula (1), and the torsion return angle is given. by subtracting from the slip angle alpha, and calculates the effective slip angle alpha _e. The reason why the effective slip angle α _e is calculated in this manner is that when the torque M _z is larger than 0, the torque acts on the tire itself so as to reduce the applied slip angle, and has a function of twisting back. . Therefore, when the torque M _z is larger than 0, as shown in FIG. 4A, the effective slip angle α _e becomes smaller than the actually applied slip angle α.

Further, the deflection coefficient q that defines the shape of the contact pressure distribution is calculated from the torque _Mz by the equation (2). The deflection coefficient q means that the ground pressure distribution in the straight traveling state with the slip angle α = 0 (see FIG. 5A) is generated, and the lateral force F _y is generated as shown in FIG. This is a parameter representing the shape of the contact pressure distribution deflected toward the front in the direction (the stepping end on the contact surface). This contact pressure distribution is represented by p (t) (t is a coordinate position normalized by the contact length when the t-axis is taken in the backward direction of travel in FIGS. 5A and 5B). Then, the shape of the ground pressure distribution p (t) is defined by the function D _gsp (t; n, q) represented by the equation (7) in FIG.
Here, the coefficient n in the function D _gsp (t; n, q) defines the contact pressure distribution of the contact surface during the generation of the lateral force, and as shown in FIG. It is a coefficient that defines the contact pressure distribution so that it is angular (the curvature increases) near the kicking end. Further, as shown in FIG. 5D, the peak position of the contact pressure distribution is set so as to move toward the stepping-end side as the coefficient q is changed from 0 to 1. Thus, the coefficient q and the coefficient n are shape defining coefficients that define the shape of the contact pressure distribution.

Furthermore, a value (x _c / l) representing the degree to which the tire center position moves to the depression end side when the lateral force F _y is generated is calculated in association with the torque M _z by the expression (3). Here, l is the contact length. Contacting define the movement of the tire center position O in such a formula (3), as shown in FIG. 5 (b), a tire center position O serving as a rotational center of the torque M _z due to the generation of the lateral force F _y This is to move to the stepping side of the ground.

Furthermore, the boundary position (l _h / l) between the sliding friction and the adhesion friction within the ground contact surface that occurs when the slip angle α is large is calculated from the equation (4). The boundary position (l _h / l) is defined as follows.
The maximum friction curves shown in FIGS. 6A to 6C are obtained by multiplying the adhesion friction coefficient _μs by the contact pressure distribution p (t). The tire tread member that is in contact with the road surface at the stepping end is gradually sheared from the road surface by the slip angle α as it moves to the kicking end, and a shearing force (adhesion friction force) is generated between the tire and the red member. The shear forces, reaching the maximum friction curve gradually increases, tire tread elements which have adhesion to the road surface Suberidashi accordance sliding friction curve obtained by multiplying the contact pressure distribution p (t) the sliding friction coefficient mu _d Sliding friction force is generated. In FIG. 6 (a), the region closer to the stepping end than the boundary position (l _h / l) is an adhesion region where the tire tread member is adhered to the road surface, and the region on the kicking side is the tire tread member relative to the road surface. It becomes the slipping area of the tire. FIG. 6B shows a state where the slip angle α is larger than the slip angle α shown in FIG. The boundary position (l _h / l) has moved to the stepping end side as compared with FIG. Further, when the slip angle α increases, as shown in FIG. 6C, sliding friction is generated from the position of the stepping end of the contact surface.

As can be seen from FIGS. 6A to 6C, the ratio between the adhesion region and the slip region varies greatly depending on the slip angle α. The lateral force F _y ′ can be calculated by integrating the frictional force in the adhesion region and the slip region, that is, the lateral force component along the tire width direction, and the moment around the tire center O is calculated. Thus, the torque M _z ′ can be calculated.
In the equations (5) and (6), the lateral force F _y ′ and the torque M _z ′ are calculated using the effective slip angle α _e by dividing into the above-mentioned adhesion region and slip region.

In formula (5), the lateral force F _y ′ is calculated by the sum of two terms (two lateral force components). The first term is an integral having an integration range of 0 to (l _h / l), and represents an adhesion lateral force component generated in the adhesion region. The second term is an integral whose integration range is (l _h / l) to 1, and represents a slip lateral force component generated in the slip region.

Further, in the equation (6), the first term is an integral having an integral range of 0 to (l _h / l), and represents a torque component generated by the adhesion lateral force component generated in the adhesion region, and the second term. Is an integral having an integral range of (l _h / l) to 1 and represents a torque component generated by a slip lateral force component generated in the slip region. In equation (6), in addition to the above two torque components, another torque component, that is, a third term is provided. The third term, A _x · (l _h / l) · tan α _e, is calculated by the amount of movement at this time and the longitudinal force of the tire, as will be described later. The torque component around the tire center O is expressed. That is, the torque M _z ′ is calculated by the sum of three components: a torque component generated by the adhesion lateral force, a torque component generated by the sliding lateral force, and a torque component generated by the longitudinal force.

Adhesive lateral force component of the first term in equation (5) is a lateral force in the adhesive region, in Formula (5), lateral transverse displacement of the belt tread element caused by the effective slip angle alpha _e bending deformation The adhesion lateral force component is calculated by expressing the state relaxed by.
The slip lateral force component of the second term is a lateral force in the slip region, and in Equation (5), the shape of the contact pressure distribution p (t) generated by the effective slip angle α _e is expressed by a function D _gsp (t; n, q). The slip lateral force component is calculated by

4 (a) to 4 (c) schematically represent the ground contact surface with the relationship between the effective slip angle α _e , the relaxed adhesion lateral force component generated by belt deformation and the longitudinal force component and the torque component. It is illustrated.
4 (a) is when the slip angle alpha is applied, showing a state in which acts on the tire itself to reduce the slip angle alpha by a torque caused by the slip angle alpha, which is the effective slip angle alpha _e . FIG. 4 (b) shows the relationship between the lateral displacement caused by the lateral bending deformation of the lateral displacement and the belt caused by the effective slip angle alpha _e. FIG. 4C shows a mechanism in which the longitudinal force distribution generated by the lateral contact force of the tire moving in the lateral direction contributes to the torque M _z ′. In FIG. 4C, M _z1 and M _z2 indicate a torque component due to an adhesion lateral force component and a torque component due to a sliding lateral force component, and M _z3 indicates a torque component due to a longitudinal force acting on the contact surface.

FIG. 7 is a processing block diagram until the slip angle α is given and the lateral force F _y ′ and torque M _z ′ are calculated based on the tire dynamic model. As can be seen from FIG. 7, the tire dynamic model according to the present invention has a lateral bending deformation of the belt, a shape change of the contact pressure distribution, and a torsional deformation of the tire when calculating the lateral force F _y ′ and the torque M _z ′. It is fed back and calculated in equations (5) and (6). Here, _the lateral force F _{y ',,} and _the torque M _z' are next to the belt used to calculate bending deformation, and the shape change and twisting of the tire deformation ground pressure distribution, the lateral force F _y and torque is applied M _z is used.

Note that the lateral force F _y ′ and torque M _z ′ calculated by the tire dynamic model calculation unit 14 do not necessarily match the applied lateral force F _y and torque M _z . However, the sequence processing performed in CP / SATP parameter calculation program 20 and _F y / _{M z} parameter calculation program 22, CP / SATP data calculating program _24, F y / _{M z} data calculation program 26, described below, the tire dynamic model in, lateral force and F _{y ',} and _the torque M _z' are calculated as the lateral force F _y and _the torque M _z is applied substantially coincides (the equilibrium of forces) manner, the lateral force F _y and the torque M _z is searched, and the lateral force and torque in a balanced state in the tire dynamic model are calculated.

Next, a description is given of functions of the CP / SATP parameter calculating program 20, and _F y / _{M z} parameter calculation program 22, CP / SATP data calculating program _24, F y / _{M z} data calculation program 26.
The CP / SATP parameter calculation program 20 is obtained using the values of parameters B to E in which “Magic Formula” is set or corrected, and is supplied to the CP / SATP parameter calculation program 20. An error between the lateral force F _y and the torque M _{z at} one degree and the lateral force F _y ′ and the torque M _z ′ is equal to or less than a predetermined value, that is, the lateral force and the torque are in a state of force balance in the tire dynamic model. This is a program for deriving the above-described linear parameters and nonlinear parameters.

FIG. 8 shows the flow of processing performed in the CP / SATP parameter calculation program 20.
Specifically, the data of the lateral force F _y and the torque M _{z at} the slip angle α = 1 degree, and the data of the contact length l and the contact width w of the tire in the non-rolling state at the load load F _z and the load load F _z . Is acquired in advance (step S100). These data are data stored in the memory 4 and are called from the memory 4. Alternatively, an instruction input from the input operation system 5 may be used.
Further, the lateral bending coefficient ε and the torsional compliance (1 / G _mz ) are initially set to predetermined values (step S102).

Next, using the initial set values of the lateral bending coefficient ε and torsional compliance (1 / G _mz ), the lateral stiffness K _y0 and the longitudinal stiffness A _x that are linear parameters are calculated by linear square regression that is a known method ( Step S104). In addition, in the ground contact surface at the slip angle α = 1 degree, when the shape of the ground contact surface is rectangular and the longitudinal force density acting on the ground contact surface is uniform and proportional to the average contact pressure, the lateral stiffness K _y0 and the longitudinal stiffness A _x is determined in a proportional relationship with the calculated value using the contact length l, the contact width w, and the load load F _z as in the following formula, and the proportional constant at this time is the lateral stiffness K _y0 and the longitudinal stiffness A _x . It is a parameter that characterizes the value.
^{_{K y0 α w · l 2/}} 2
A _x Ｆ Fz · l / 2

Here, since the nonlinear parameter is set to a predetermined value, a normal equation for the lateral stiffness K _y0 and the longitudinal stiffness A _x that are linear parameters can be determined, and the linear parameter is obtained by solving this normal equation. The lateral stiffness K _y0 and the longitudinal stiffness A _x can be uniquely calculated. Specifically, the non-linear parameters set so that the lateral forces F _y ′ and M _z ′ calculated by the equations (5) and (6) optimally return to the supplied lateral forces F _y and M _z. Is used to create a normal equation for the lateral stiffness K _y0 and the longitudinal stiffness A _x , which are linear parameters, from the formula of the sum of squared residuals, and the lateral stiffness K _y0 and the longitudinal stiffness A _x are calculated by solving this normal equation. Here, the normal equation is an equation in which an equation for determining the residual sum of squares is partially differentiated with respect to the respective linear parameters of the lateral stiffness K _y0 and the longitudinal stiffness A _x , and the partial differential value is 0, and the number of linear parameters This is an equation related to the linear parameter to be created.

Next, the obtained lateral force F _y and torque M _z data, the calculated linear parameters, and the initialized nonlinear parameters are given to the tire dynamic model calculation program 14. In the tire dynamic model calculation program 14, the lateral force F _y ′ at the slip angle α = 1 degree and the data of the lateral force F _y and the torque M _z and the flow of the processing block diagram shown in FIG. Torque M _z ′ is calculated. In this case, since the slip angle α is 1 degree, the ground pressure distribution coefficient n of the ground pressure distribution at the slip angle α = 0 degree is fixed, and the deflection coefficient q is set to 0. Further, the boundary position (l _h / l) between the adhesion region and the slip region is set to 1. That is, there is no slip area on the ground contact surface, and all are adhesion areas. Therefore, the slip lateral force component of the second term in the equations (5) and (6) and the torque component generated by this slip lateral force line segment Becomes 0. Further, since the contact length l is given as the measurement data, the contact length and regression exponential function of the applied load F _z the contact length l which is measured instead of the effective contact length l _e is a linear parameter in FIG A function, that is, a function having load load F _z dependency is used.

  Here, the contact width w is the width in the lateral direction of the tread member of the tire, and the tread member is provided with a tire groove that forms a tread pattern. For this reason, since the actual ground contact area actually in contact with the road surface is different from the total ground contact area of the tread member, the ground contact width w corrected using the ratio of actual ground contact area / total ground contact area is used.

Next, the lateral force F _y ′ and the torque M _z ′, which are the corresponding calculation data of the lateral force and torque at the slip angle α = 1 degree calculated by the tire dynamic model calculation program 14, are stored in the CP / SATP parameter calculation program 20. Using the calculated data of the lateral force F _y ′ and torque M _z ′ and the data of the lateral force F _y and torque M _{z at} the slip angle α = 1 degree, it is expressed by the following formula (8). A compound square residual sum _Qc is calculated (step S106).

Here, N is the condition setting number of the load condition in which the load load changes, and i is an integer of 1 to N. Further, _{g f} and _{g m} are when a lateral force _{F y} and dispersing sigma _f ² and sigma _m ² of the lateral force _{F y} and _the torque _{M z} in the N load condition for the data of the torque _{M z,} the following formula in a coefficient expressed, a weighting factor used to obtain the combined sum of squared residuals Q _c.
g _f = 1 / σ _f ²
g _m = 1 / σ _m ²
In other words, the composite square residual sum _Qc is obtained by weighting and adding the square residual sums of the lateral force and torque with the reciprocal of the variance, which is information of measurement data variation, as a weighting coefficient.

Thus, a multiplied by the weighting factor g _f the residual sum of squares of the lateral force value of F _{y 'calculated} data with the value of the lateral force F _y measurement data, the value of the torque M _z of the measurement data The composite square residual sum is calculated by adding the residual square sum with the value of the torque M _z ′ of the calculated data and the weighted coefficient g _m .
Here, the combined sum of squares is used to optimize the lateral force F _y ′ and torque M _z ′ under a plurality of load conditions in the calculation of the nonlinear parameter, and optimally for the corresponding lateral force F _y and torque M _z. This is because they match.

Further, it is determined whether or not the composite square residual sum is equal to or less than a predetermined value (step S108).
If it is determined that they have not converged, the previously set nonlinear parameters of the transverse bending coefficient ε and torsional compliance (1 / G _mz ) are adjusted (step S110). The adjustment of the nonlinear parameter is performed according to, for example, the Newton-Raphson method. Specifically, an equation relating the matrix and the adjustment amount of the nonlinear parameter is obtained by performing a second-order partial differentiation of the composite square residual sum with respect to the nonlinear parameter, and the equation is solved with respect to the adjustment amount. Then, the adjustment amount of the nonlinear parameter is calculated. This calculation method is described in detail in Japanese Patent Application No. 2001-242059 (Japanese Patent Laid-Open No. 2003-57134) filed by the applicant of the present application.

Each time this non-linear parameter is adjusted, linear least-squares regression (step S104) with respect to the linear parameter and calculation of a composite square residual sum (step S106) are performed to obtain a composite square residual sum according to equation (8). Then, the adjustment of the nonlinear parameter is repeated until the composite square residual sum becomes a predetermined value or less.
If the combined sum of squared residuals becomes equal to or less than a predetermined value, a linear least squares regression lateral stiffness K _y0 calculated _in, the lateral bending coefficient is around rigid A _x and nonlinear parameters epsilon, torsional compliance with (1 / G _mz) Parameters (Step S112). The determined parameters are stored in the memory 20.
The flow of calculation of the linear parameter and the nonlinear parameter at the slip angle α = 1 degree using the tire dynamic model performed by the CP / SATP parameter calculation program 20 is as described above.

The F _y / M _z parameter calculation program 22 changes the slip angle α to, for example, 0 to 20 degrees, and the lateral force F _y ′ and the torque M _z ′ when the adhesion region and the slip region exist on the contact surface are This is a part for calculating the above-mentioned linear parameter and non-linear parameter so as to coincide with the lateral force F _y and torque M _z calculated from the values of the parameters B to E of “Magic Formula”.

FIG. 9 shows the flow of processing performed in the F _y / M _z parameter calculation program 22.
Specifically, as shown in FIG. 9, the characteristic curves of the lateral force F _y and torque M _z calculated from the values of parameters “B” to “E” of “Magic Formula” by variously changing the slip angle at a constant load are obtained. Obtain (step S200).
Further, the mechanical element parameters are set by reading the lateral bending coefficient ε and the torsional compliance (1 / G _mz ) obtained by the CP / SATP parameter calculation program 20 and stored in the memory 4 (step S202).
Further, the remaining nonlinear parameters, coefficient n, lateral stiffness coefficient (K _y0 / μ _s ), coefficient C _q , and movement coefficient C _xc are initialized to predetermined values (step S204).

Next, the linear least squares regression using the "Magic Formula" nonlinear parameters characteristic curve and is initially set in the calculated lateral force F _y and _the torque M _z from the values of the parameters B~E in (step S206). Specifically, normal equations relating to linear parameters such as the lateral stiffness K _y0 and the longitudinal stiffness A _x are created, and by solving this normal equation, the lateral stiffness K _y0 , the longitudinal stiffness A _x , the sliding friction coefficient μ _d , the effective grounding The length l _e is calculated. That is, linear least square regression is performed. Here, the normal equation is an equation relating to linear parameters created by the number of linear parameters in which the residual sum of squares is partially differentiated with respect to each of the above linear parameters and the partial differential value is zero.
The linear parameters calculated using the nonlinear parameters and normal equations thus initialized and the data of the characteristic curves of the lateral force F _y and the torque M _z are given to the tire dynamic model calculation program 14. With this application, the lateral force F _y ′ and torque M _z ′ at each slip angle α are calculated according to the flow of the block diagram of FIG.

Next, using the calculation data of the lateral force F _y ′ and torque M _z ′ and the characteristic curve data of the lateral force F _y and torque M _{z at} the applied slip angle α, the above equation (8) is used. in represented by calculating the combined sum of squared residuals _{Q c} (step S208). In this case, N in equation (8) is the condition setting number of the slip angle α to be given. The weighting coefficients g _f and g _{m at} this time are obtained from the variance of the lateral force F _y and the torque M _z under N slip angle conditions.

Thus, the residual sum of squares between the value of the lateral force F _y calculated from the values of the parameters B to E of the “Magic Formula” and the value of the lateral force F _y ′ calculated by the tire dynamic model calculation program 14 Multiplied by the weighting coefficient g _f , the value of the torque M _z calculated from the values of the parameters B to E of the “Magic Formula”, and the value of the torque M _z ′ calculated by the tire dynamic model calculation program 14 Is added to the sum of the residual squared squares multiplied by the weighting factor g _m to calculate the composite squared residual sum. Here, similar to the above-described case, the composite square residual sum is used to calculate the lateral force F _y ′ and the torque M _z ′ under the conditions of a plurality of slip angles in the calculation of the nonlinear parameter as the lateral force F _y and the torque M. _This is because it is made to coincide with each of _z simultaneously.
It is determined whether or not the composite square residual sum is equal to or less than a predetermined value (step S210).
If it is determined that it has not converged, the nonlinear parameter initially set in step S204 is adjusted (step S212). The adjustment of the nonlinear parameter is performed according to, for example, the Newton-Raphson method.

In the embodiment shown in FIG. 9, the lateral bending coefficient ε and the torsional compliance (1 / G _mz ), which are nonlinear parameters, are used as the nonlinear parameters determined by reading from the memory 4. The nonlinear parameters of the bending coefficient ε and the torsional compliance (1 / G _mz ) may be set as initial parameters as undetermined parameters as with other nonlinear parameters. However, if the non-linear parameters of the transverse bending coefficient ε and the torsional compliance (1 / G _mz ) are calculated by the CP / SATP parameter calculating program 20, the calculated lateral bending coefficient ε and the torsional compliance (1 / G _mz ) is preferably used. This is because the value of the dynamic element parameter obtained by the CP / SATP parameter calculation program 20 and the value of the dynamic element parameter obtained by the F _y / M _z parameter calculation program 22 are matched.

This non-linear parameter is adjusted until it is determined that it converges in step S210. Each time this adjustment is performed, linear least square regression (step S206) and calculation of a composite square residual sum (step S208) regarding the linear parameter are performed. Then, the composite residual sum of squares according to the above equation (8) is obtained. Then, the non-linear parameter adjustment is performed until the composite square residual sum becomes a predetermined value or less. When the composite square residual sum becomes a predetermined value or less, nonlinear parameters such as lateral stiffness K _y0 and longitudinal stiffness A _x calculated by linear least square regression are determined (step S214), and these dynamic element parameters are stored in memory 4 To remember.
The above is the flow of calculation of the linear parameter and the nonlinear parameter at each slip angle α using the tire dynamic model performed by the F _y / M _z parameter calculation program 22.

The CP / SATP data calculation program 24 calculates the lateral force F _y ′ and torque M _z ′ under different load conditions at the slip angle α = 1 degree, the linear parameters calculated by the CP / SATP parameter calculation program 20, and This is a part that is calculated using nonlinear parameters.
FIG. 10 shows the flow of processing performed in the CP / SATP data calculation program 24.

Specifically, as shown in FIG. 10, the CP / SATP data calculation program 24 first reads and sets the linear parameters and nonlinear parameters calculated by the CP / SATP parameter calculation program 20 from the memory 4 (step S300). ).
Further, the lateral force _{F y} and _the torque _{M z} under the applied load F _z is initialized (step S302).
Thereafter, the linear parameter and the non-linear parameter are given to the tire dynamic model calculation program 14 together with the slip angle α = 1 degree and the initially set lateral force F _y and torque M _z . In the tire dynamic model calculation program 14, the applied linear and non-linear parameters and the lateral force F _y and torque M _{z that} are set in _advance are used to obtain the lateral force F according to the equations (5) and (6) in FIG. _y ′ and torque M _z ′ are calculated (step S304).

In this case, since the slip angle α = 1 degree, the ground pressure distribution coefficient n of the ground pressure distribution at the slip angle α = 0 degree is fixed, and the deflection coefficient q is set to zero. Further, the boundary position (l _h / l) between the adhesion area and the sliding area is set to 1. That is, there is no slip area on the ground contact surface, and all are adhesion areas. Therefore, the slip lateral force component of the second term in the equations (5) and (6) and the torque component generated by this slip lateral force line segment Becomes 0. When the ground contact length l is given, this ground contact length l is used as a default value instead of the effective ground contact length l _e which is a linear parameter in FIG.

The thus calculated lateral force F _y ′ and torque M _z ′ with the slip angle α = 1 degree are returned to the CP / SATP data calculation program 24. The CP / SATP data calculation program 24 is a composite square of the set values of the lateral force F _y and torque M _z given to the tire dynamic model calculation program 14 and the calculated values of the calculated lateral force F _y ′ and torque M _z ′. The residual sum is calculated according to equation (8) (step S306).

Next, it is determined whether or not the composite square residual sum is equal to or less than a predetermined value (step S308).
If it is determined that it has not converged, the set values of the lateral force _Fy and the torque _Mz set previously are adjusted (step S310). The adjusted lateral force F _y and torque M _z are again applied to the tire dynamic model calculation program 14 together with the linear parameter and the nonlinear parameter.
Thus, the set values of the lateral force F _y and the torque M _z are adjusted until the composite square residual sum becomes equal to or less than a predetermined value and converges. This set value is adjusted, for example, according to the Newton-Raphson method described above. Thus, the convergent lateral force F _y ′ and torque M _z ′ are determined (step S312).

Furthermore, the conditions of _the applied load _{F z} is changed (step S314). Each time the applied load F _z is changed, the lateral force F _y and the torque M _z are initialized (step S302), and the lateral force F _y 'and the torque M _z ' are calculated using the set values (step S304). ) A composite square residual sum is calculated (step S306), and the convergence of this composite square residual sum is determined (step S308).
Thus, _the applied load _{F z} is repeatedly changed until the predetermined load (step S316). The lateral force F _y ′ and torque M _z ′ are calculated every time the applied load F _z is changed, and the converging lateral force F _y ′ and torque M _z ′ are determined. The determined lateral force F _y ′ and torque M _z ′ are stored in the memory 20.
In this way, we obtain a curve which depends on the applied load F _z CP and SATP is the lateral force and the torque at the slip angle alpha = 1 degree.

The F _y / M _z data calculation program 26 is a linear parameter calculated by the F _y / M _z parameter calculation program 22 for the lateral force F _y ′ and torque M _z ′ at a plurality of slip angles α at a predetermined load. And it is a part which calculates using a nonlinear parameter.
FIG. 11 shows the flow of processing performed in the F _y / M _z data calculation program 26.
First, the F _y / M _z data calculation program 26 reads and sets the linear parameter and the nonlinear parameter calculated by the F _y / M _z parameter calculation program 22 from the memory 4 (step S400).
Further, the lateral force _{F y} and _the torque _{M z} under the applied load F _z is initialized (step S402).
Thereafter, linear and non-linear parameters and the initial lateral force F _y and torque M _z are given to the tire dynamic model calculation program 14 together with the set slip angle α = Δα. In the tire dynamic model 14, the applied linear and nonlinear parameters, the lateral force F _y and the torque M _{z that} are set in _advance are used, and the lateral force F _y ′ and the torque M _z according to the equations (5) and (6). 'Is calculated (step S404).

The lateral force F _y ′ and torque M _z ′ calculated in this way are returned to the F _y / M _z data calculation program 26. The F _y / M _z data calculation program 26 calculates the set values of the lateral force F _y and the torque M _z given to the tire dynamic model calculation program 14 and the calculated values of the calculated lateral force F _y ′ and torque M _z ′. The composite sum of squared residuals is calculated according to equation (8) (step S406).
Next, it is determined whether or not the calculated composite square residual sum is equal to or smaller than a predetermined value (step S408).
If it is determined that it has not converged, the previously set lateral force F _y and torque M _z set values are adjusted (step S410). The adjusted lateral force F _y and torque M _z , linear parameters, and nonlinear parameters are provided to the tire dynamic model calculation program 14 again.
In this way, the set values of the lateral force F _y and the torque M _z are adjusted until the composite square residual sum becomes a predetermined value or less and converges. This set value is adjusted, for example, according to the Newton-Raphson method described above. Thus, the lateral force F _y ′ and the torque M _z ′ are determined (step S412).

Next, it is determined whether or not the slip angle α is equal to or smaller than a predetermined slip angle (step S416).
When it is determined that the slip angle α is equal to or smaller than the predetermined slip angle, the condition of the slip angle α is changed (α → α + Δα) (step S414). Then, initial values of the lateral force F _y and torque M _{z at} the changed slip angle α are set (step S402), the lateral force F _y ′ and torque M _z ′ are calculated (step S404), and the composite square residual is calculated. The sum is calculated (step S406), and the convergence of this composite square residual sum is determined (step S408).
Thus, the slip angle α is repeatedly changed until the predetermined slip angle is reached (step S416). The lateral force F _y ′ and torque M _z ′ are calculated every time the slip angle is changed, and the converging lateral force F _y ′ and torque M _z ′ are determined. The determined lateral force F _y ′ and torque M _z ′ are stored in the memory 4.
In this way, a characteristic curve of lateral force and torque depending on the slip angle α is obtained.
The above is the description of the configuration of the device 1.

Using such a device 1, a vehicle is designed according to the flow shown in FIG.
First, in the setting program 9, data of vehicle specifications and values of parameters B to E of “Magic Formula” of tires are set (steps S600 and S602). These settings may be set by calling predetermined data from the memory 4, or may be input by an input operation system 5.
Next, the vehicle travel simulation program 10 generates a vehicle model based on the set vehicle specification data. For example, a vehicle model based on a mechanism analysis model is generated (step S604). Further, a traveling simulation condition is set (step S606). Different driving simulation conditions are set according to the performance to be evaluated. For example, when the performance evaluation is durability evaluation, profile data such as the vehicle traveling speed and road surface roughness is set as the traveling condition. In the case of the emergency avoidance performance evaluation, data such as vehicle traveling speed and steering angle, actual road surface profile data, and the like are set as traveling conditions.
Further, under the traveling simulation condition, a traveling simulation is performed using the vehicle model while calculating the lateral force and torque from the set values of the parameters “B” to “E” of “Magic Formula” (step S608).
The running simulation result is stored in the memory 4.

  Next, in the integration / management program 13, the driving simulation result is called from the memory 4, the performance evaluation data of the set performance index is calculated (step S 610), and the performance evaluation data is set to a preset target value. It is determined whether or not it is satisfied (step S612). If the performance evaluation data does not satisfy the target value, it is determined whether or not the vehicle specification data can be corrected (step S614). If the result is affirmative, the vehicle specification data is corrected (step S616). The corrected vehicle specification data is sent to the vehicle travel simulation program 10 again, a vehicle model is generated again, the values of the parameters “B” to “E” of “Magic Formula” are given to the vehicle model, and the vehicle travel simulation is performed. Done.

In this way, the vehicle specification data is corrected until the performance evaluation data satisfies the target value, and the performance evaluation is repeated each time.
If the performance evaluation data does not satisfy the target value and the vehicle specification data cannot be corrected any more, for example, if the vehicle wheelbase exceeds the regulated length, the vehicle specification data Instead, the values of the parameters B to E of “Magic Formula” are corrected (step S618). The method for correcting the values of parameters B to E is not particularly limited. For example, the values of parameters B to E are sequentially changed within a predetermined width.
As described above, the tire characteristic correcting process is performed after at least one process of correcting the vehicle specification data and performing the driving simulation and the vehicle performance evaluation, so that the target performance cannot be achieved with the vehicle specifications. Can be efficiently achieved with the tire characteristics, and the vehicle including the tire can be efficiently designed.

Next, the dynamic element parameters in the tire dynamic model are derived using the values of the parameters B to E to which the correction has been made (step S620). Specifically, the “Magic Formula” data / parameter calculation program 11 calculates a characteristic curve representing the slip angle dependency and load dependency of the lateral force and torque, and the characteristic curve is stored in the memory 4. Since the values of parameters B to E are set for each of a plurality of load conditions, a slip angle-dependent characteristic curve is generated for each load condition. Next, the mechanical element parameters in the tire dynamic model are extracted by the CP / SATP parameter calculation program 24, and the remaining dynamic element parameters in the tire dynamic model are derived by the F _y / M _z parameter calculation program 26. (Step S620). The derived dynamic element parameters are stored in the memory 4 as dynamic element parameters corresponding to the modified parameters BE.

In the derivation of the dynamic element parameters, depending on the degree of convergence (step S210 in FIG. 9), the characteristic curves represented by the values of parameters B to E and the characteristic curves represented by the dynamic element parameters do not completely match. There are many cases. In this case, after step S620, the characteristic curve of the slip force dependency of the lateral force and torque represented by the dynamic element parameters is calculated by the F _y / M _z data calculation program 26, and the calculated characteristic curve is “ It is preferable to supply to the “Magic Formula” data parameter calculation program 11 to derive the values of the parameters B to E. This is because the values of the derived parameters B to E are very close to the values of the parameters B to E modified in step S618 and are guaranteed to be realizable by the tire dynamic model.

Next, the corrected values of parameters B to E are given to the vehicle model, and a running simulation is performed. In this manner, the values of the parameters B to E of “Magic Formula” are corrected until the determination in step S612 is positive, and a driving simulation using the vehicle model is performed each time, and performance evaluation is performed.
If the determination in step S612 is affirmative, the data of the vehicle specification is determined, and further, the values of the parameters “B” to “E” of the “Magic Formula” and the values of the dynamic element parameters of the tire are determined as the required characteristics of the tire (step). S622).
Based on the data of the vehicle specifications thus determined, detailed design of components such as suspensions is performed, and the tire is provided with the determined mechanical element parameters as the required characteristics of the tire, and the tire manufacturer is notified of the tire specifications. Is placed (step S624).

  Thus, since the required characteristics of a tire can be realized by a tire dynamic model, it can be said that the required characteristics can be realized by structural design and material design of the tire. Therefore, a feasible tire that conforms to the determined vehicle specification data can be ordered from the tire manufacturer. In addition, tire characteristics can be determined without using a complicated dynamic model of the tires. Therefore, by using a database of material characteristics of tire rubber members in combination, tire design characteristics are reasonable even for vehicle designers. It can be judged whether it is a thing. This contributes to shortening the vehicle design period. In addition, since the required characteristics are presented according to the dynamic element parameters of the tire, a smooth development system with the vehicle designer and the tire designer is possible, and development efficiency is dramatically improved. In particular, since the performance of the vehicle can be evaluated from the stage where there is no actual tire, performance evaluation such as emergency avoidance performance can be performed from an early stage, and the performance of the vehicle can be improved in a short time.

  As described above, the apparatus 1 is characterized in that the dynamic element parameters are derived using the tire dynamic model. The validity of the derivation of the dynamic element parameters using the tire dynamic model will be described below.

FIG. 13A shows a characteristic curve l ₁ of CP load dependence calculated by using the composite square residual sum in the CP / SATP parameter calculation program 20 and the CP / SAT data calculation program 24, and the composite square residual. The parameter is calculated without using the difference sum, and the characteristic curve l ₂ of the load dependency of CP calculated without using the composite square residual sum using this parameter, and the measured lateral force F _y (in the figure, “○” is plotted). FIG. 13B shows the SATP load dependency characteristic curve l ₃ calculated using the composite square residual sum as in the characteristic curve l ₁ and the composite square residual as in the characteristic curve l _2. sum and SATP load dependency characteristic curve l ₄ of which is calculated without using, _(in the figure, plotted with "○") measured torque M _z indicates a, a.
FIG. 13 (a), the from (b) As is apparent, curve l _1, l ₃ calculated by using the combined sum of squared residuals is characteristic curve l ₂ calculated without using the combined sum of squared residuals, compared to l ₄ it can be seen that correspond to very good measurement values.

14A and 14B show the characteristic curves l ₅ and l of the load dependence of CP and SATP created using the characteristic curves l ₁ and l ₂ and the Fiala model which is a known tire dynamic model. ₆ . The known Fiala model is a model in which the effective slip angle α _e in the tire dynamic model in the present invention and the longitudinal force component applied to the SATP by the longitudinal force acting on the contact surface are not considered.
As is apparent from FIGS. 14A and 14B, the characteristic curves l ₁ and l ₂ calculated using the tire dynamic model of the present invention are characteristic curves l ₅ and l calculated using the known Fiala model. _It can be seen that it corresponds to the measured value very well compared to FIG.

FIG. 15 shows a characteristic curve ₁₇ representing the load dependence of SATP calculated by the CP / SAT data calculation program 24. The characteristic curve ₁₇ corresponds very well to the measured values. Further, in this characteristic curve ₁₇ , the longitudinal force torque component (one-dot chain line) in SATP is shown. As can be seen from FIG. 15, the longitudinal force torque component shows approximately one-third contribution in SATP. Thus, in the present invention, the torque characteristic curve can be decomposed and expressed, and the tire cornering characteristic can be analyzed in detail using the tire dynamic model.

FIGS. 16A and 16B show characteristic curves l ₈ and l ₉ representing the slip angle dependence of the lateral force F _y ′ and torque M _z ′ calculated by the F _y / M _z data calculation program 26. Is shown. The characteristic curves l ₈ and l ₉ correspond very well to the measured values.
In FIG. 16 (a), the characteristic curve l ₈ is divided into the first term adhesion lateral force component (dotted line) and the second term slip lateral force component (dashed line) in the above equation (5). Yes. In FIG. 16B, the lateral force torque component generated by the lateral force (adhesion lateral force component + slip lateral force component) of the first term and the second term in the above formula (6) and the longitudinal force torque generated by the longitudinal force. It is divided into components and displayed. As described above, in the present invention, the characteristic curves of the lateral force and the torque can be decomposed and expressed, and the cornering characteristics of the tire can be analyzed in detail.
FIG. 16C is a diagram illustrating the state of the contact pressure distribution represented by the function D _gsp (t; n, q) when the parameters are calculated by the F _y / M _z parameter calculation program 22. It can be seen that as the generated self-aligning torque _Mz increases, the peak of the contact pressure is biased toward the stepping-in end, and the contact surface moves toward the kick-out side.

Thus, in the present invention, the tire characteristic curve can be accurately expressed using the derived dynamic element parameters. For this reason, the dynamic element parameters are derived from the values of the parameters “B” to “E” of the “Magic Formula” for a predetermined tire, and a part of the result is changed to determine how the characteristic curve changes. Sensitivity can be found.
As described above, the tire dynamic model used in the present invention can reproduce the characteristic curves of the lateral force and the torque more accurately than the conventional one.

  The vehicle design method including the tire of the present invention has been described in detail above, but the present invention is not limited to the above-described embodiment, and various improvements and modifications can be made without departing from the spirit of the present invention. Of course it is good. Although the above design method has mainly described the relationship between the tire lateral force generated on the tire rotation shaft and the vehicle design, in the present invention, the vehicle is designed by using the same tire dynamic model in the tire longitudinal direction. It is clear that this method can be used for simulations related to vehicle design in general.

It is a block diagram of the apparatus of one Example which implements the design method of the vehicle containing the tire of this invention. It is a figure explaining the tire dynamic model used in the design method of vehicles including the tire of the present invention. It is another figure explaining the tire dynamic model used in the design method of vehicles including the tire of the present invention. (A)-(c) is another figure explaining the tire dynamic model used in the design method of the vehicle containing the tire of this invention. (A)-(d) is another figure explaining the tire dynamic model used in the design method of the vehicle containing the tire of this invention. (A)-(c) is another figure explaining the tire dynamic model used in the design method of the vehicle containing the tire of this invention. It is a processing block diagram until it calculates a lateral force and torque in the tire dynamic model used in the design method of vehicles including the tire of the present invention. It is a flowchart which shows the flow of one process implemented in the design method of the vehicle containing the tire of this invention. It is a flowchart which shows the flow of the other process implemented in the design method of the vehicle containing the tire of this invention. It is a flowchart which shows the flow of one process implemented in the design method of the vehicle containing the tire of this invention. It is a flowchart which shows the flow of the other process implemented in the design method of the vehicle containing the tire of this invention. It is a flowchart which shows the flow of an example of the design method of the vehicle containing the tire of this invention. (A) And (b) is a figure which shows an example of the output result obtained by the design method of the vehicle containing the tire of this invention. (A) And (b) is a figure which shows the other example of the output result obtained by the design method of the vehicle containing the tire of this invention. It is a figure which shows the other example of the output result obtained by the design method of the vehicle containing the tire of this invention. (A)-(c) is a figure which shows the other example of the output result obtained by the design method of the vehicle containing the tire of this invention. It is a figure which shows the design method of the conventional vehicle.

Explanation of symbols

1 Device 2 CPU
3 Bus 4 Memory 5 Input operation system 6 Interface 7 Output device 8 Program group 9 Setting program 10 Vehicle running simulation program 11 “Magic Formula” data / parameter calculation program 12 Tire dynamic model program group 13 Integration / management program 14 Tire dynamic model calculation Program 20 CP / SATP parameter calculation program 22 F _y / M _z parameter calculation program 24 CP / SATP data calculation program 26 F _y / M _z data calculation program

Claims (7)

By performing vehicle running simulation using information on characteristic curves representing slip ratio dependence of force or torque acting on the tire rotation axis based on shearing force acting on the road surface and information on vehicle specifications A vehicle design method including tires for designing a vehicle having a desired vehicle performance,
A model creation step for creating a vehicle model using information on vehicle specifications;
When the characteristic curve is approximated by a nonlinear approximation formula, a value of an approximate formula parameter that defines the nonlinear approximation formula is assigned to the vehicle model, a running simulation is performed under a predetermined running condition, and the result of the running simulation is used. A performance evaluation step for evaluating the performance of the vehicle,
In the performance evaluation, when the vehicle model does not satisfy the target performance, the vehicle specification information is corrected, a vehicle model is generated using the corrected vehicle specification information, and the travel simulation and the vehicle performance are generated. Vehicle specification correction step for evaluation,
In the performance evaluation, when the vehicle model does not satisfy a predetermined target performance, the value of the approximate expression parameter is corrected, the corrected approximate expression parameter value is assigned to the vehicle model, and a travel simulation is performed. The vehicle performance is evaluated using the results of the above, and a characteristic curve is calculated from the nonlinear approximate expression defined by the corrected approximate expression parameter value. From this characteristic curve, a plurality of tire dynamic element parameters are used. Tire characteristic correction step for deriving a value of a tire dynamic element parameter that defines the characteristic curve based on a tire dynamic model
A tire characteristic determining step for determining, as a tire required characteristic, the value of the tire dynamic element parameter derived corresponding to the corrected value of the approximate expression parameter when the vehicle model satisfies a target performance;
The method for designing a vehicle including a tire, wherein the tire characteristic correcting step is performed after the vehicle specification correcting step is performed at least once.   The method for designing a vehicle including a tire according to claim 1, wherein the characteristic curve is a curve representing the slip angle dependency of a lateral force generated on a tire shaft and a self-aligning torque when a slip angle is given. The tire dynamic model calculates a lateral force and generates a lateral force torque component caused by a lateral force acting on the tire contact surface and a longitudinal force torque component caused by a longitudinal force acting on the tire contact surface. The method for designing a vehicle including a tire according to claim 2, wherein the model is a model for calculating self-aligning torque separately.   In the tire characteristic correction step, when deriving the value of the tire dynamic element parameter, the sum of squared residuals of the characteristic curve of the lateral force and the corresponding curve of the lateral force calculated by the tire dynamic model, and the self The sum of squares of the characteristic curve of the aligning torque and the corresponding curve of the self-aligning torque calculated by the tire dynamic model is a value obtained by weighted addition using a weighting coefficient, and the weighting coefficient is as follows: Compound square residual using coefficients obtained from information on variation of values that vary depending on slip angles of the characteristic curves of the lateral force and the self-aligning torque calculated by the value of the approximate expression parameter Deriving tire dynamic element parameter values that represent characteristics during tire cornering so that the sum is less than or equal to a predetermined value. Vehicle design method, including tire according to claim 2 or 3.   In the tire characteristic correction step, when deriving the value of the tire dynamic element parameter from the characteristic curve based on the tire dynamic model, it is given by the torsional deformation of the tire generated by self-aligning torque. The method for designing a vehicle including a tire according to any one of claims 2 to 4, wherein a value of a tire dynamic element parameter is derived using an effective slip angle in which the slip angle is corrected.   2. The tire dynamic element parameter derived in the tire characteristic correcting step includes an adhesion friction coefficient between a tire tread member and a road surface, and a shape defining coefficient defining a sliding friction coefficient and a shape of a contact pressure distribution. A design method of a vehicle including the tire according to any one of?   In the tire characteristic correcting step, the adhesion friction is determined using at least one of a stiffness parameter for tire shear deformation, a stiffness parameter for tire lateral bending deformation, and a stiffness parameter for tire torsional deformation, which is obtained in advance. The vehicle design method including a tire according to claim 6, wherein a coefficient, the sliding friction coefficient, and the shape defining coefficient are derived.

Images