JP2006201089A - Model characteristic generation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique capable of creating a proper behavior simulation model with a load applied onto a measured object taken into consideration. <P>SOLUTION: In a behavior simulation model creating system 100, a component force measuring instrument 110 oscillates the measured object by a prescribed input wave, and detects the load and a displacement of the measured object under the oscillated condition. A computing part 130 finds a hexa-axial matrix operation expression as to a relation between the load and the displacement as the behavior simulation model, using a method of least squares, a neural network or the like, based on the load and the displacement of the measured object. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for creating a behavior simulation model of an object to be measured.

Conventionally, in vehicle development and the like, a number of techniques for simulating behavior using parts as objects to be measured have been proposed. For example, Patent Document 1 discloses a technique for analyzing system behavior based on a spring constant obtained from a system in which a spring / damping element having an arbitrary frequency characteristic and a spring / damping element having an arbitrary nonlinearity coexist. Has been.
JP 2004-101304 A JP 2003-287461 A JP 2004-132901 A JP 2003-139669 A Japanese Patent Laid-Open No. 10-260999

  However, in the spring / damping model employed in the above-described technology, the shape of the object to be measured and the characteristics of the member such as rubber used for the object to be measured cannot be appropriately modeled, preventing accurate simulation. There was a problem of being.

  For example, consider the case of simulating the behavior of a suspension bush (hereinafter referred to as “bush”), which is a rubber component used in a vehicle. FIG. 1 shows a cross-sectional view when the bush is crossed in the axial direction. In the bush 500 shown in the figure, a shaft hole 502 is formed at the center, and a tickling 504 that is two cavities facing each other across the X axis is formed outside the shaft hole 502. Here, the X axis is defined in the right direction of FIG.

  When the behavior of the bush 500 is simulated, the actual characteristics may not be accurately expressed in the nonlinear region. 2 shows an input characteristic in the X direction (solid direction) of the bush 500 in FIG. 1, FIG. 3 shows an input characteristic in the Y axis direction (a direction crossing the curl), and FIG. 4 shows a 45 degree direction (with the X axis). The simulation results of the input characteristics in the middle direction of the Y axis are shown. As is clear from a comparison between FIG. 2 and FIG. 3, the straight direction of the bush 500 is softer and more easily deformed than the solid direction. The simulation result of the input characteristic in the intermediate direction shown in FIG. 4 is obtained by combining the input characteristic in the solid direction and the input characteristic in the fast direction. However, in practice, the input characteristic in the intermediate direction is an intermediate characteristic between the input characteristic in the solid direction and the input characteristic in the fast direction in the nonlinear region, and is not accurately simulated. That is, in the actual measurement, the curve indicating the input characteristics in the intermediate direction appears in the middle of the input characteristics in the solid direction and the input characteristics in the straight direction, but in the simulation result, it appears in a softer direction than the straight direction.

Further, in the simulation of the rotational acceleration of the handle, the degree of coincidence between the simulation result and the actual measurement value may be reduced. FIG. 5 shows the simulation results and actual measurement values of the first bush (referred to as before bushing countermeasures), and FIG. 6 shows the simulation results and actual measurement values of the second bushing (referred to as after bushing countermeasures) with different spring constants. In any of these simulations, the bush is modeled by a spring element, and the spring constant obtained by a static load is used as the spring constant. Before the countermeasure against bushing, the degree of coincidence between the simulation result and the actual measurement value is high as shown in FIG. 5, whereas after the countermeasure against bushing, the degree of coincidence between the simulation result and the actual measurement value is low as shown in FIG. Then, as shown in FIG. 7, after the bush countermeasure, the degree of coincidence between the simulation result and the actual measurement value is increased when the spring constant is about twice the actual value. This is because the displacement at the time of vibration changes when the spring constant of the bush changes due to the displacement dependency of the spring constant of the bush, and further, the actual operation of the bush can be simulated as a result of the change of the spring constant at the time of vibration. It is speculated that it is not. As described above, since the displacement dependency of the characteristics of the bush varies depending on the individuality of the bush, it is difficult to specify the spring constant at the load that vibrates from the spring constant at the static load.

  FIG. 8 shows measured values of the acceleration frequency characteristics of the acceleration of the handle when the handle equipped with the bush is vibrated in a state where a load (hereinafter referred to as preload) is applied in advance. In the actual measurement value, the peak (corresponding to the resonance point) is measured with a difference between the case where the preload is the first value (solid line) and the case where the preload is the second value (dotted line) larger than the first value. FIG. 9 shows a simulation result under the same conditions. In the simulation, the peak of the frequency characteristic of acceleration does not change according to the amount of preload, and it is difficult to reproduce the actual resonance characteristic of the bush at the time of preload.

  Furthermore, there is a case where simulation of a bush having a complicated shape such as a to-collect bush cannot be appropriately performed. FIG. 10 is a cross-sectional view of a bush 500, which is an example of a toe collect bush, vertically cut in the axial direction. In FIG. 5, the hatched portion is the bush 500. The bush 500 is fitted between an inner cylinder 514 disposed on the outer periphery of the shaft 512 and an outer cylinder 516. As shown in the actually measured elastic main shaft in FIG. 11, the elastic main shaft in the bush 500 is displaced in accordance with the load, and the to-collect amount fluctuates accordingly (in FIG. 11, the second load is the second load). It is a value larger than the load of 1). However, in the simulation, the elastic main shaft is not displaced according to the load while maintaining the design specification elastic main shaft in FIG. 11, and the to-collect amount does not change. As described above, if the bush is modeled as a spring element using the spring constant for the static load, it is difficult to execute a simulation that matches the actual movement of the bush.

  An object of the present invention is to provide a technique that enables the creation of a behavior simulation model that captures the relationship between the load and displacement applied to an object to be measured such as a bush and matches the actual operation of the object to be measured. And

  In order to solve the above problems, the present invention employs the following means. That is, the present invention relates to the relationship between the vibration step for exciting the object to be measured with a predetermined input wave, the detection step for detecting the load and displacement of the object to be measured, and the load and displacement. And an arithmetic step for obtaining a six-axis matrix arithmetic expression composed of a direction parallel to each axis of the three-dimensional space defined by the axis and the Z axis and a rotation direction around each axis.

  By adopting such a configuration, the relationship between the load applied to the object to be measured and the displacement of the object to be measured due to the load can be expressed by a six-axis matrix arithmetic expression, taking into account the load applied to the object to be measured. Appropriate model characteristics can be generated.

Further, the present invention is directed to the external force of the object to be measured with respect to six axes including a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotation direction around each axis. A model characteristic generation method for obtaining a coefficient matrix indicating a mechanical characteristic, and among the six axes, any four axes including three axes parallel to an X axis, a Y axis, and a Z axis are vibrated with a predetermined input wave. A coefficient indicating a relationship between the vibration step to be performed, a measurement step for measuring the load in the six-axis direction and the displacement in the four-axis direction during the vibration, and the load in the six-axis direction and the displacement in the four-axis direction And an analysis step for obtaining a matrix.
With such a configuration, the displacement due to the vibration to the four axes and the resulting load in the six axes are measured, and a coefficient matrix indicating the relationship between the displacement and the load can be obtained.
The model characteristic generation method combines coefficient matrices obtained in each of the three combinations obtained by combining any of the six axes including three axes parallel to the X axis, the Y axis, and the Z axis. And a synthesis step for obtaining a coefficient matrix indicating the relationship between the load in the six-axis direction and the displacement in the six-axis direction. With such a configuration, it is possible to obtain a coefficient matrix indicating the relationship between the load in the 6-axis direction and the displacement in the 6-axis direction by combining the measurement results obtained by exciting the 4-axis.
In addition, the present invention relates to the object to be measured, six axes comprising a direction parallel to the respective axes in a three-dimensional space defined by the X axis, the Y axis, and the Z axis, and a rotational direction around the respective axes. The 6-axis direction load and the 6-axis direction displacement are measured during the excitation, and a coefficient matrix indicating the relationship between the 6-axis direction load and the 6-axis direction displacement is obtained. Thus, such a model characteristic generation method may be used in which a coefficient matrix indicating a dynamic characteristic with respect to the external force of the object to be measured is obtained. Further, the present invention may be a model analysis device that realizes such a function. According to the present invention, with respect to the object to be measured, six axes comprising a direction parallel to the respective axes in a three-dimensional space defined by the X axis, the Y axis, and the Z axis, and a rotational direction around the respective axes. The six-axis direction load and the six-axis direction displacement are measured, and a coefficient matrix indicating the relationship between the six-axis direction load and the six-axis direction displacement can be obtained. .
In the model characteristic generation method, the design object is constituted by a shape model on a computer, the shape model is divided into finite elements, and the excitation step and the measurement step are based on physical property information of the object to be measured. It may be performed on the shape model in which a mechanical relationship between the finite elements is defined.
With this configuration, when the vibration state of the measurement object is simulated by the finite element method, the load applied to the measurement object derived from the simulation result and the displacement of the measurement object due to the load Can be expressed by a six-axis matrix calculation expression, and appropriate model characteristics can be generated in consideration of the displacement dependency of the characteristics of the object to be measured.

  In the model characteristic generation method, the object to be measured may be an elastic member.

  With such a configuration, it is possible to appropriately generate a behavior simulation model of an elastic member in which the relationship between the load and the displacement is not uniform. Furthermore, in the model characteristic generation method, the elastic member may be a bush.

Further, the present invention relates to a vibration means for vibrating a measured object with a predetermined input wave, a measuring means for measuring the load and displacement of the measured object, and the relationship between the load and displacement in the X axis. Analyzing means for obtaining a 6-axis matrix operation formula consisting of a direction parallel to each axis of the three-dimensional space defined by the Y axis and the Z axis and a rotation direction around each axis. It may be a model characteristic generation device.
Further, the present invention is directed to the external force of the object to be measured with respect to six axes including a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotation direction around each axis. A model characteristic generating apparatus for obtaining a coefficient matrix indicating a mechanical characteristic, and among the six axes, any four axes including three axes parallel to the X axis, the Y axis, and the Z axis are vibrated with a predetermined input wave. Excitation means, load measurement means for measuring the load in the six axes during the vibration, displacement measurement means for measuring the displacement in the four axes during the vibration, the load in the six axes, and the It may be a model characteristic generation device having analysis means for obtaining a coefficient matrix indicating a relationship with displacement in the four axial directions.
Further, the present invention may be a program that causes a computer to realize any one of the above functions. Here, the computer uses the X axis, the Y axis among the six axes including the direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and the rotation direction around each axis. An excitation means for exciting any four axes including a shaft and three axes parallel to the Z axis with a predetermined input wave, a load measurement means for measuring a load in the six-axis direction during the excitation, and the excitation It may be connectable to an apparatus provided with a displacement measuring means for measuring the displacement in the four axial directions during shaking.
Further, the computer constitutes the design object by a shape model on the computer, the shape model is divided into finite elements, and the excitation step and the measurement step are performed based on the physical property information of the object to be measured. It may be performed on the shape model in which the mechanical relationship between elements is defined.
The present invention may also be a computer-readable recording medium that records such a program.
Here, the computer-readable recording medium refers to a recording medium that accumulates information such as data and programs by electrical, magnetic, optical, mechanical, or chemical action and can be read from the computer. . Examples of such a recording medium that can be removed from the computer include a flexible disk, a magneto-optical disk, a CD-ROM, a CD-R / W, a DVD, a DAT, an 8 mm tape, and a memory card.
Further, there are a hard disk, a ROM (read only memory), and the like as a recording medium fixed to the computer.

  According to the present invention, it is possible to create an appropriate behavior simulation model in consideration of the load applied to the object to be measured.

  A behavior simulation model creation system according to the best mode for carrying out the present invention (hereinafter referred to as “embodiment”) will be described below with reference to the drawings. In addition, the structure of the following embodiment is an illustration and this invention is not limited to the structure of embodiment.

  FIG. 12 shows a configuration diagram of the first behavior simulation model creation system according to the embodiment of the present invention. The behavior simulation model creation system 100 shown in FIG. 12 is a matrix calculation formula for a behavior simulation model of a rubber bush or the like to be measured (hereinafter referred to as “measurement part”, which corresponds to the measurement object of the present invention). Is what you want.

This behavior simulation model creation system 100 includes a vibration unit 112 (this corresponds to the vibration unit of the present invention), and performs vibration on a rubber bush or the like to be measured. A component force measuring device 110 that detects a load and a displacement, a load / displacement data storage unit 120 that stores load and displacement data of the measurement component detected by the component force measuring device 110, and a load and displacement of the measurement component And a matrix operation expression storage unit 140 that stores the matrix operation expression obtained by the operation unit 130.
Among these, the load / displacement data storage unit 120, the calculation unit 130, and the matrix calculation formula storage unit 140 are realized by a computer program executed on a computer including a CPU, a memory, a hard disk, and the like (not shown). The component force measuring device 110 is connected to the CPU via an interface (not shown), and is controlled by a computer program executed on the CPU to perform vibration and measurement.

  Hereinafter, the creation of a behavior simulation model by the behavior simulation model creation system 100 and the operation of the behavior simulation of the measurement component following this will be described. FIG. 13 shows a flowchart of behavior simulation model creation and behavior simulation operations.

  First, a measurement part to be subjected to behavior simulation is selected and placed on the vibration unit 112 in the component force measuring device 110 (S101).

When the measurement component is arranged in the vibration unit 112, the component force measuring apparatus 110 drives the vibration unit 112 to vibrate the measurement component with a predetermined input random wave (S102). The vibration unit 112 vibrates the measurement component in the directions of these three axes by moving in the directions of three axes (X axis, Y axis, and Z axis) orthogonal to each other determined for the measurement component. FIG. 14 shows an example of loads in the X-axis, Y-axis, and Z-axis directions applied to the measurement component. As shown in the figure, Fx, Fy, and Fz, which are loads in the X-axis, Y-axis, and Z-axis directions, are random waves. Further, the excitation unit 112 rotates the measurement component by rotating about the axis in the vertical direction as the rotation axis. That is, in the excitation by the excitation unit 112, a total of four axes of excitation in the direction of three axes and excitation in the rotational direction around the vertical direction are performed on the measurement component. It should be noted that the vibration can be performed on the four axes at substantially the same time, or can be performed by shifting the time for each of the four axes.

  In the present embodiment, the direction of the measurement component that is the object to be measured is changed, so that the vibration is performed when the X axis is in the vertical direction, when the Y axis is in the vertical direction, and when the Z axis is in the vertical direction. By doing so, the X-axis, Y-axis, and Z-axis of the measurement component are excited in the rotational direction around the axis. As a result, the measurement component is vibrated on the X axis, the Y axis, the Z axis, and a total of six axes in the rotational direction around these axes.

  The component force measuring device 110 (corresponding to the load measuring means and the displacement measuring means of the present invention) is a state in which the vibration pickup unit 112 is vibrated, and a built-in acceleration pickup (not shown), strain sensor (Not shown) or the like is used to detect the respective loads of the six axes of the measurement component and the respective displacement data of the six axes of the measurement component due to the load. The load and displacement data of the measurement component detected by the component force measuring device 110 is sent to and stored in the load / displacement data storage unit 120 (S103).

Next, the calculation unit 130 reads the load data and the displacement data stored in the load / displacement data storage unit 120, and uses, for example, the least square method to calculate the relationship between the load and the displacement as a behavior simulation model. An axis matrix calculation formula is obtained (S104). In the implementation of the present invention, the procedure for obtaining the matrix arithmetic expression is not limited to the least square method. That is, instead of the calculation by the least square method, the calculation may be performed by a neural network.
Here, a load vector F which is a 6 × 1 matrix corresponding to the load of the measurement component, a motion state vector D which is a 42 × 1 matrix corresponding to the displacement of the measurement component, and the relationship between the load and the displacement A coefficient matrix H, which is a 6-axis by 42-column 6-axis matrix, and a constant vector b, which is a 6-by-1 matrix corresponding to friction and the like in the measurement component,
(Equation 1)
F _{(6 × 1)} = H _{(6 × 42)}・ D _{(42 × 1)} + b _{(6 × 1)}
There is a relationship. Here, the load vector F uses Fx, Fy, Fz, etc., which are loads in the X-axis, Y-axis, and Z-axis directions,
(Equation 2)
F _{(6 × 1)} = (Fx Fy Fz ...) ^T
The motion state vector D is represented by the displacement δx, the speed obtained by differentiating the displacement δx, the acceleration obtained by differentiating the speed, and the power of the displacement δx,
(Equation 3)
D _{(42 × 1)} = (δx ... δx '... δx''... δx ² ... δx ³ ... δx ⁴ ... δx ⁵ ...)
It is represented by Here, δx is a displacement in the X direction, δx ′ is a velocity in the X direction, δx ″ is an acceleration in the X direction, δx ^{2 and the} like are terms of a power of displacement. The arithmetic unit 130 can obtain a coefficient matrix H that is a six-axis matrix based on the relationship of Equation 1 described above. By obtaining such a relational expression, compared to the conventional case where a measuring part such as a bush is modeled by a simple spring constant of a static load, the motion state vector and load including a higher-order term for the displacement are compared. With the coefficient matrix H. Also, in this relational expression, the relationship between the displacement and the load is conventionally obtained for each of the six axes, whereas the relationship between the anisotropic elements, for example, the load parallel to the X direction and the X direction A relationship with a displacement other than the parallel direction (a displacement parallel to the Y and Z directions, a displacement in the rotational direction around the X, Y, and Z axes) is quantified by the coefficient matrix H.

  Hereinafter, an example of derivation of the coefficient matrix H that is a six-axis matrix regarding the relationship between the load and the displacement as the behavior simulation model will be described. In the following, for the sake of simplicity of explanation, an example will be described in which the coefficient matrix H is a 6 × 6 matrix, the motion state vector D is a 6 × 1 matrix, and the constant vector b does not exist.

  FIG. 15 shows a bush which is a measurement part. In the bush 300 shown in the figure, the load and displacement are obtained for a total of six axes in the X-axis, Y-axis, and Z-axis directions and the rotational directions around the X-axis, Y-axis, and Z-axis. A matrix H is obtained.

  Specifically, the load is set to Fx, Fy, Fz, Mx, My, and Mz for the X-axis, Y-axis, and Z-axis directions and the six rotation axes around the X-axis, Y-axis, and Z-axis, respectively. Assuming that the displacements are Dx, Dy, Dz, Ax, Ay, and Az, respectively, the equation for obtaining the motion state vector D based on the above equation 1 is a coefficient matrix H as shown in FIG. As shown in FIG. 16A, the equation for obtaining the load vector F is a further inverse matrix of the inverse matrix of the coefficient matrix H, that is, an expression using the coefficient matrix H. It is represented by Accordingly, a plurality of measured values of load and displacement are obtained for the X-axis, Y-axis, and Z-axis directions, and the six rotation axes around the X-axis, Y-axis, and Z-axis. The coefficient matrix H may be obtained by the least square method, a neural network, or the like from the above equation or the equation in FIG.

  However, as described above, the excitation unit 112 in the component force measuring device 110 can apply excitation in the X-axis, Y-axis, and Z-axis directions and rotation around the vertical direction in one excitation. Only a total of 4 axes of vibration can be performed. For this reason, by changing the direction of the bushing 300 as the measurement component so that the vibration is performed three times, the X-axis, Y-axis, and Z-axis directions, and these X-axis, Y-axis, and Z-axis The load and displacement are obtained with respect to the six axes with the rotation direction around.

  That is, first, as shown in FIG. 17, the bush 300 is disposed in the vibration portion 112 so that the X axis is in the vertical direction, and the direction parallel to the X axis, the Y axis, and the Z axis of the bush, Four axes in the direction of rotation are vibrated. During this excitation, the loads Fx, Fy, Fz, Mx, My, and Mz in the respective directions of the six axes, the X-axis, Y-axis, and Z-axis directions of the six axes, and the rotational directions around the X-axis The four-axis displacements Dx, Dy, Dz, and Ax are detected by the component force measuring device 110. Then, as shown in FIG. 18, the calculation unit 130 converts these loads Fx, Fy, Fz, and the like into an equation using a coefficient matrix of 4 rows and 6 columns composed of the first row to the fourth row of the coefficient matrix H. By substituting Mx, My, Mz and displacements Dx, Dy, Dz, Ax, the coefficients in the first to fourth lines of the coefficient matrix H are obtained by the least square method, a neural network, or the like (this processing is executed). The calculation unit 130 corresponds to an analysis unit).

Next, as shown in FIG. 19, the bush 300 is arranged in the vibration unit 112 so that the Y axis is in the vertical direction, and is vibrated. By this vibration, the loads Fx, Fy, Fz, Mx, My, and Mz in the respective directions of the six axes, the X-axis, Y-axis, and Z-axis directions of the six axes, and the rotational directions around the Y-axis The four-axis displacements Dx, Dy, Dz, and Ay are detected by the component force measuring device 110. Then, as shown in FIG. 20, the calculation unit 130 converts these loads Fx into an equation using a coefficient matrix of 4 rows and 6 columns composed of the first row to the third row and the fifth row of the coefficient matrix H. , Fy, Fz, Mx, My, Mz and displacements Dx, Dy, Dz, Ay are substituted to obtain the coefficients of the first to third lines and the fifth line of the coefficient matrix H. Obtained by least square method, neural network, etc. Note that the coefficients in the first to third lines of the coefficient matrix H obtained here are substantially the same as the coefficients obtained when the X-axis is in the vertical direction, in the range of measurement errors.

  Further, as shown in FIG. 21, the bush 300 is arranged in the vibration unit 112 so that the Z-axis is in the vertical direction, and is vibrated. By this vibration, the loads Fx, Fy, Fz, Mx, My, and Mz in the respective directions of the six axes, the X-axis, Y-axis, and Z-axis directions of the six axes, and the rotational directions around the Z-axis The four-axis displacements Dx, Dy, Dz, and Az are detected by the component force measuring device 110. Then, as shown in FIG. 22, the calculation unit 130 converts these loads Fx into an equation using a coefficient matrix of 4 rows and 6 columns composed of the first row to the third row and the sixth row of the coefficient matrix H. , Fy, Fz, Mx, My, Mz and displacements Dx, Dy, Dz, Az are substituted to obtain the coefficients of the first to third lines and the sixth line of the coefficient matrix H as the least square method. Calculate with a neural network. Note that the coefficients in the first to third lines of the coefficient matrix H obtained here are substantially the same as the coefficients obtained when the X axis is in the vertical direction and the Y axis is in the vertical direction, in the range of measurement errors. It is.

  By adding higher order terms, differential terms, etc. and increasing the number of coefficients constituting the coefficient matrix H, it is possible to derive a more accurate coefficient matrix H that represents a behavior simulation model.

  Returning to FIG. 13, the description will be continued. The six-axis matrix calculation formula regarding the relationship between the load and the displacement as the behavior simulation model obtained by the calculation unit 130 is sent to and stored in the matrix calculation formula storage unit 140 (S105, this processing is executed). The calculation unit 130 corresponds to a combining unit).

  Thereafter, a simulation is performed using the six-axis matrix arithmetic expression. That is, the simulation apparatus (not shown) reads out the six-axis matrix arithmetic expression stored in the matrix arithmetic expression storage unit 140, sets it to the user-defined element format of the general-purpose finite element method program (S106), and this general-purpose finite element method program To perform a simulation of the structure including the measurement component (S107). Specifically, the bush portion is described by a spring element having the characteristics of the coefficient matrix obtained based on the analysis of the measurement result, and the vehicle portion other than the bush (the main body side and the suspension connected to the main body side by the bush) Side) is divided into finite elements, and the rigidity of the entire vehicle is simulated.

In this way, the behavior simulation model creation system 100 vibrates the measurement component with a predetermined input random wave, detects the load and displacement of the measurement component, and determines six axes about the relationship between the load and displacement. Is calculated as a behavior simulation model. Therefore, it is possible to create an appropriate behavior simulation model that reflects the load applied to the measurement component and the relationship between the load and the displacement more accurately than in the past. In particular, when the measurement component is an elastic member such as rubber, it is easily deformed by a load. Therefore, a more appropriate behavior simulation model can be created by considering the relationship between the load and the displacement. .
That is, the behavior simulation model creation system 100 can model the measurement component by obtaining the relationship between the motion state vector including a higher-order term for the displacement and the load using the coefficient matrix H. The behavior simulation model creation system 100 also has different characteristics, for example, a load parallel to the X direction and a displacement other than a direction parallel to the X direction (displacement parallel to the Y and Z directions, X, Y, and Z axes). And the measurement component can be modeled by obtaining the relationship with the coefficient matrix H. The same applies to rotation in the direction parallel to the Y and Z directions and around the X, Y and Z axes.

In the above-described embodiment, vibration in a direction parallel to the three coordinate axes of the bush, the X axis, the Y axis, and the Z axis, and one fixed axis, for example, a vertical direction (for example, the X axis direction of the bush) The rotation of the surrounding rotation was performed. During this vibration, the loads Fx, Fy, Fz, Mx, My, and Mz in the respective directions of the six axes, the directions parallel to the X axis, the Y axis, and the Z axis among the six axes, and the vertical direction (for example, Four-axis displacements Dx, Dy, Dz, Ax consisting of rotation around the bush (X-axis direction) were detected by the component force measuring device 110. And the coefficient of the corresponding part of the coefficient matrix H was calculated | required from the measurement result.
Similarly, the direction of the bush was changed, and instead of such excitation around the X axis, excitation was performed around the Y axis, and the coefficient of the corresponding portion of the coefficient matrix H was obtained. Further, the coefficient of the corresponding portion of the coefficient matrix H was obtained by changing the direction of the bush and oscillating around the Z axis instead of oscillating around the Y axis. In this way, measurement was performed on three different sets of four axes, a partial matrix of the coefficient matrix H was obtained, and the coefficient matrix H was obtained by combining these partial matrices.
However, the implementation of the present invention is not limited to a procedure that combines such partial direction excitation and measurement. For example, excitation in directions parallel to the X axis, Y axis, and Z axis, excitation in parallel to these two axes having a predetermined inclination with these axes, and vertical direction (for example, coincident with the X axis direction) Is applied to the bush through the excitation unit, and the displacement and acceleration in the six-axis directions (X-axis, Y-axis, Z-axis, and other components parallel to the other two axes and the vertical axis) Find the relationship. Furthermore, these 6-axis components are transformed into components parallel to the X, Y, and Z axes and components around the X, Y, and Z axes, and the relationship between the displacement and acceleration. The coefficient matrix H may be obtained from
In addition, for example, 6-axis vibration (X-axis, Y-axis, Z-axis vibration and vibration around each axis) is executed almost simultaneously, and the displacement and acceleration in 6-axis direction are measured. The coefficient matrix H may be obtained by one measurement.
Therefore, instead of being fixed to a single axis in the rotational direction, it is only necessary to generate a rotational motion in which the rotational components around the X, Y, and Z axes vary pseudo-randomly with time. Therefore, for example, in the excitation as shown in FIG. 17, the rotation axis of the rotary motion shown as the excitation around the X axis (vertical axis) in the figure is not fixed around the X axis, and the rotation axis May be moved in a random direction indicated by (δx, δy, δz).
Therefore, the excitation unit may be provided with a first rotation unit that performs excitation around the rotation axis and a second rotation unit that rotates the rotation axis itself around a predetermined fulcrum. With such a configuration, such a rotation shaft may be swung around the fulcrum. Such a rotating shaft may be swung around a predetermined fulcrum, and the tip of the rotating shaft may be moved circularly. Furthermore, a third power source that varies the inclination of the turning surface (or the swinging surface formed by the rotation shaft of the first rotating unit when swinging) when rotating around the fulcrum may be provided. What is necessary is just to comprise a 3rd power source so that the rotating shaft of a said 1st rotation part may be rotated in the direction substantially orthogonal to a turning surface.
In this case, the displacement may be measured by providing a six-axis strain sensor on the bush in advance. In this case, since the direction of the bush itself changes every moment due to the rotation of the first rotation unit, the second rotation unit, and / or the third rotation unit, the measurement directions of the respective strain sensors are respectively changed. The timing that coincides with the X axis, Y axis, and Z axis is detected from the rotating motion of the bush (the rotating state of the first rotating portion to the third rotating portion), and the direction parallel to each axis at that timing What is necessary is just to measure the displacement around each axis. In addition, the timing when the measurement direction of each acceleration pickup coincides with the X, Y, and Z axes is detected from the rotational movement of the bush, and the acceleration parallel to each axis and the acceleration around each axis is measured by the timing. do it.
Incidentally, the behavior simulation model creation system 100 described above actually vibrates the measurement component and detects the load and displacement of the measurement component. However, the load and displacement of the measurement component may be estimated by simulation. good.

  FIG. 23 shows a configuration diagram of a second behavior simulation model creation system according to the embodiment of the present invention. Similar to the simulation model creation system 100 described above, the behavior simulation model creation system 200 shown in FIG. 23 obtains a behavior simulation model of a measurement part such as a rubber bush as a measurement object as a matrix operation expression. .

  The behavior simulation model creation system 200 performs simulation of the vibration state of the measurement part, and stores the vibration state simulation unit 210 that detects the estimated load and the estimated displacement, and the characteristic data of the material of the measurement part as physical property information. Material property database (DB) 212, load / displacement data storage unit 220 for storing estimated load and estimated displacement data of the measurement component detected by the vibration state simulation unit 210, and estimated load and estimated displacement of the measurement component And a storage unit 240 for storing the matrix operation expression obtained by the operation unit 230.

  Hereinafter, the creation of a behavior simulation model by the behavior simulation model creation system 200 and the subsequent behavior simulation operation of the measurement component will be described. FIG. 24 shows a flowchart of behavior simulation model creation and behavior simulation operations.

  First, the vibration state simulation unit 210 creates three-dimensional (3D) CAD (Computer Aided Design) data of the shape of a measurement part to be subjected to behavior simulation in accordance with an operator's operation instruction or the like (S201). Next, the vibration state simulation unit 210 executes commercially available software for creating a finite element method model. By executing this software, the vibration state simulation unit 210 divides the analysis object represented by the 3D CAD data into meshes, and further creates a finite element method model for each mesh (S202).

  Next, the vibration state simulation unit 210 executes commercially available software for finite element analysis. By executing this software, the vibration state simulation unit 210 uses the characteristic data corresponding to the material of the measurement part and the data of the finite element method model stored in the material characteristic database 212 to generate a predetermined input random wave. A simulation of the state in which the measurement component is vibrated is performed (S203).

  Further, the vibration state simulation unit 210, based on the result of the vibration state simulation, estimates the load of each of the six axes of the measurement component (hereinafter referred to as “estimated load”) and the measurement based on the load. Data of estimated values (hereinafter referred to as “estimated displacements”) of displacements of the six axes of the part is detected. The estimated load and estimated displacement data of the measurement component detected by the vibration state simulation unit 210 is sent to and stored in the load / displacement data 220 (S204).

  Thereafter, the same operations as those after S104 in FIG. 13 are performed. That is, the calculation unit 230 reads the estimated load data and the estimated displacement data stored in the load / displacement data storage unit 220, and uses the least square method, a neural network, or the like to estimate the estimated load and the estimated displacement as a behavior simulation model. A six-axis matrix calculation formula for the relationship is obtained (S205). A specific method for deriving a 6-axis matrix calculation formula is the same as S104 in FIG. The six-axis matrix calculation formula regarding the relationship between the estimated load and the estimated displacement as the behavior simulation model obtained by the calculation unit 230 is sent to and stored in the matrix calculation formula storage unit 240 (S206).

  Next, the simulation apparatus (not shown) reads out the 6-axis matrix arithmetic expression stored in the matrix arithmetic expression storage unit 240 and sets it in the user-defined element format of the general-purpose finite element method program (S207). The program is executed, and a simulation for the structure including the measurement component is performed (S208).

As described above, the behavior simulation model creation system 200 performs the simulation of the vibration state of the measurement component, detects the estimated load and the estimated displacement of the measurement component, and has six axes about the relationship between the estimated load and the estimated displacement. Is calculated as a behavior simulation model. Accordingly, even when there is no measurement part, as in the behavior simulation model creation system 100, it is possible to create an appropriate behavior simulation model in consideration of the load applied to the measurement part.
Further, since many bushes are used in the vehicle, the amount of calculation can be reduced and the CPU load can be reduced as compared with the case where the entire vehicle including the bushes is divided into finite elements for simulation.

  Since an appropriate behavior simulation model is created in consideration of the load applied to the object to be measured, a more appropriate behavior simulation model is created especially when the object to be measured is an elastic member that is easily deformed by the load. It becomes possible.

3 is an upper cross-sectional view of a bush 500. FIG. It is a figure which shows the input characteristic of the solid direction of the bush. It is a figure which shows the input characteristic of the right direction of the bush. It is a figure which shows the input characteristic of the intermediate direction of the bush. It is a figure which shows the simulation result and measured value of the rotational acceleration of the handle | steering-wheel before a bush countermeasure. It is a figure which shows the simulation result and measured value of the rotational acceleration of the handle | steering-wheel after a bush countermeasure. It is a figure which shows the simulation result and measured value of the rotational acceleration of a handle | steering-wheel corresponding to each spring constant after a bush countermeasure. It is a figure which shows the 1st simulation result of the rotational acceleration of a steering wheel when loaded beforehand. It is a figure which shows the 2nd simulation result of the rotational acceleration of a steering wheel when loaded beforehand. It is side sectional drawing of the bush 500 with a complicated shape. It is a figure which shows a response | compatibility with a design specification elastic principal axis and a measurement elastic principal axis. 1 is a configuration diagram of a first behavior simulation model creation system 100 according to an embodiment of the present invention. FIG. It is a flowchart of operation | movement of creation of the behavior simulation model by the 1st behavior simulation model creation system 100 which concerns on embodiment of this invention, and subsequent behavior simulation. It is a figure which shows the time passage of the load added to a measurement component. It is an external view of the bush 300 which is a measurement component. It is a figure which shows the type | formula for calculating | requiring the motion state vector D and the load vector F. FIG. It is a figure which shows the 1st arrangement | positioning state of the bush. It is a figure which shows the type | formula for calculating | requiring the motion state vector D in a 1st arrangement | positioning state. It is a figure which shows the 2nd arrangement | positioning state of the bush. It is a figure which shows the type | formula for calculating | requiring the motion state vector D in a 2nd arrangement | positioning state. It is a figure which shows the 3rd arrangement | positioning state of the bush. It is a figure which shows the type | formula for calculating | requiring the motion state vector D in a 3rd arrangement | positioning state. It is a block diagram of the 2nd behavior simulation model creation system 200 which concerns on embodiment of this invention. It is a flowchart of operation | movement of creation of the behavior simulation model by the 2nd behavior simulation model creation system 200 which concerns on embodiment of this invention, and subsequent behavior simulation.

Explanation of symbols

100, 200 Behavior simulation model creation system 110 Component force measurement device 112 Excitation unit 120, 220 Load / displacement data storage unit 130, 230 Calculation unit 140, 240 Matrix calculation expression storage unit 210 Excitation state simulation unit 212 Material property DB 300 bush

Claims (16)

An excitation step for exciting the object to be measured with a predetermined input wave;
A measuring step for measuring the load and displacement of the object to be measured;
As for the relationship between the load and the displacement, a six-axis matrix calculation formula is obtained which includes a direction parallel to the respective axes in the three-dimensional space defined by the X, Y, and Z axes and a rotational direction around each of the axes. A model characteristic generation method comprising: an analysis step. A coefficient indicating the mechanical characteristics with respect to the external force of the object to be measured with respect to six axes including a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotation direction around each axis. A model property generation method for obtaining a matrix,
Of the six axes, an excitation step of exciting any four axes including three axes parallel to the X axis, the Y axis, and the Z axis with a predetermined input wave;
A measurement step of measuring the load in the six-axis direction and the displacement in the four-axis direction during the vibration;
An analysis step of obtaining a coefficient matrix indicating a relationship between the load in the six-axis direction and the displacement in the four-axis direction. The model characteristic generation method according to claim 2,
Among the six axes, the load in the six-axis direction is obtained by combining coefficient matrices obtained in each of the three combinations formed by combining any four axes including three axes parallel to the X-axis, the Y-axis, and the Z-axis. And a synthesis step for obtaining a coefficient matrix indicating the relationship between the displacement in the six-axis direction. In the model characteristic generation method according to any one of claims 1 to 3,
The design object is composed of a shape model on a computer, the shape model is divided into finite elements, and the excitation step and the measurement step are performed based on physical property information of the object to be measured. A model characteristic generation method, which is executed on the shape model for which a relationship is defined. In the model characteristic generation method according to any one of claims 1 to 4,
A method for generating a model characteristic, wherein the object to be measured is an elastic member. The model characteristic generation method according to claim 5,
The method for generating a model characteristic, wherein the elastic member is a bush. Vibration means for vibrating the object to be measured with a predetermined input wave;
Measuring means for measuring the load and displacement of the object to be measured;
Regarding the relationship between the load and the displacement, a six-axis matrix operation formula consisting of a direction parallel to the respective axes in a three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotational direction around the respective axes. A model characteristic generating apparatus comprising: an analyzing means to be obtained. A coefficient indicating the mechanical characteristics with respect to the external force of the object to be measured with respect to six axes including a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotation direction around each axis. A model property generation device for obtaining a matrix,
Among the six axes, vibration means for exciting any four axes including three axes parallel to the X axis, the Y axis, and the Z axis with a predetermined input wave;
Measuring means for measuring the load in the six-axis directions during the excitation;
Measuring means for measuring the displacement in the four-axis direction during the excitation;
An apparatus for generating a model characteristic, comprising: an analysis unit for obtaining a coefficient matrix indicating a relationship between the load in the six-axis direction and the displacement in the four-axis direction. The model characteristic generation device according to claim 8, wherein
Among the six axes, the load in the six-axis direction is obtained by combining coefficient matrices obtained in each of the three combinations formed by combining any four axes including three axes parallel to the X-axis, the Y-axis, and the Z-axis. And a model characteristic generation apparatus characterized by further comprising synthesis means for obtaining a coefficient matrix indicating the relationship between the displacement in the six-axis direction. In the model characteristic generation device according to any one of claims 7 to 9,
The design object is composed of a shape model on a computer, the shape model is divided into finite elements, and the vibration means and the measurement means are mechanically connected between the finite elements based on physical property information of the object to be measured. A model characteristic generating apparatus for measuring the load and displacement from the shape model in which a relationship is defined. An excitation step for exciting a measured object with a predetermined input wave to a computer;
A measuring step for measuring the load and displacement of the object to be measured;
As for the relationship between the load and the displacement, a six-axis matrix calculation formula is obtained which includes a direction parallel to the respective axes in the three-dimensional space defined by the X, Y, and Z axes and a rotational direction around each of the axes. A model characteristic generation program characterized by causing an analysis step to be executed. The computer has a mechanical characteristic with respect to the external force of the object to be measured with respect to six axes composed of a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotation direction around each axis. Is a model characteristic generation program for obtaining a coefficient matrix indicating
Of the six axes, an excitation step of exciting any four axes including three axes parallel to the X axis, the Y axis, and the Z axis with a predetermined input wave;
A measurement step of measuring the load in the six-axis direction and the displacement in the four-axis direction during the vibration;
An analysis step of obtaining a coefficient matrix indicating a relationship between the load in the six-axis direction and the displacement in the four-axis direction. In the model characteristic generation program according to claim 12,
Among the six axes, the load in the six-axis direction is obtained by combining coefficient matrices obtained in each of the three combinations formed by combining any four axes including three axes parallel to the X-axis, the Y-axis, and the Z-axis. And a synthesis step for obtaining a coefficient matrix indicating the relationship between the displacement in the six-axis direction. In the model characteristic generation program according to any one of claims 11 to 13,
The design object is composed of a shape model on a computer, the shape model is divided into finite elements, and the excitation step and the measurement step are performed based on physical property information of the object to be measured. A model characteristic generation program which is executed for the shape model in which a relationship is defined. Exciting to the object to be measured six axes comprising a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis, and a rotational direction around each axis; ,
A measurement step of measuring the load in the six-axis direction and the displacement in the six-axis direction during the vibration;
A coefficient matrix indicating a mechanical characteristic with respect to an external force of the object to be measured, comprising: an analysis step for obtaining a coefficient matrix indicating a relationship between the load in the six-axis direction and the displacement in the six-axis direction during the vibration. The model characteristic generation method to be obtained. Means for exciting the object to be measured with six axes composed of a direction parallel to each axis of the three-dimensional space defined by the X axis, the Y axis, and the Z axis and a rotational direction around each axis; ,
Means for measuring the load in the six-axis direction during the vibration;
Measuring means for measuring displacement in the six-axis directions during the excitation;
A model characteristic for obtaining a coefficient matrix indicating a mechanical characteristic with respect to an external force of the object to be measured, comprising: an analysis unit for obtaining a coefficient matrix indicating a relationship between the load in the six-axis direction and the displacement in the six-axis direction Generator.

Images