KR20200038362A - Optimal design method for conical rubber spring of railway vehicles and optimized conical rubber spring - Google Patents

Abstract

The present invention relates to an optimized design method of a conical rubber spring for a railway vehicle and to a conical rubber spring designed thereby. More specifically, provided are the optimized design method of a conical rubber spring for a railway vehicle, which optimizes the curvature of each stage of a conical rubber spring used as a suspension device for a railway vehicle to reduce localization of the strain, and improves load-deflection characteristics and rigidity, and the conical rubber spring designed thereby.

Description

{Optimal design method for conical rubber spring of railway vehicles and optimized conical rubber spring}

The present invention relates to an optimized design method of a conical rubber spring for a railway vehicle and a conical rubber spring designed thereby, more specifically, to optimize the curvature of each stage of a conical rubber spring used as a suspension device for a railway vehicle to localize the strain. The present invention relates to an optimized design method of a conical rubber spring for a railway vehicle and a conical rubber spring designed thereby to reduce and improve load-sag characteristics and rigidity.

In general, a suspension means a device that improves ride comfort and stability by alleviating shock or vibration received from a road surface while driving, and in the case of a railroad car, a primary suspension device that connects a wheel and a truck frame, and a truck frame and truck support Or it consists of a secondary suspension that connects the body.

Among them, a conical rubber spring, which is a primary suspension component, is a rubber-type axial spring having rigidity in three axes (vertical, longitudinal, and transverse).

The conical rubber spring plays an important role in the driving stability and ride comfort of the vehicle, and relieves static and dynamic loads caused by the movement of the vehicle to prevent damage to vehicle parts and tracks and minimize vibration and noise transmitted to passengers and cargo. It plays a role to let.

The primary suspension device of a railway vehicle connects a bogie and a wheel, and stabilizes the movement of the vehicle by equalizing the vertical load acting on the wheel, and reduces the external force due to the irregularity of the track. In addition, it is a key component that has the greatest influence on the critical speed of the vehicle and improves driving stability.

As shown in FIG. 1, the conventional general conical rubber spring 100 is made of a laminated rubber obtained by mutually adhering rubber and metal in a concentric shape, and has a shape in which a plurality of conical shapes are inserted, and the outside is attached to the axle box. The inside of the conical shape is fixed to the bogie frame, and is laminated with a steel core and three rubber layers: an outer layer, an intermediate layer, and an inner layer. It has three-way stiffness in the front-rear, left-right, and up-down directions, and grooves are formed in the left-right direction to have flexible stiffness characteristics compared to the front-rear direction.

Figures 2 and 3 (a) to (c) are the logarithmic strain distribution of the rubber section consisting of three rubber layers in the analysis of vertical stiffness, longitudinal (length) stiffness and lateral stiffness for tolerance and full load conditions, respectively. As a result of evaluating the strain analysis results with the von-Mises component of logarithmic strain, it was confirmed that the maximum strain was greater in the longitudinal and transverse analysis with severe loading conditions than in the vertical stiffness analysis. .

In addition, in the tolerance load of FIG. 2, the maximum strain tends to increase in the order of vertical, transverse, and longitudinal directions, but the difference in the maximum strain in the transverse direction and the longitudinal direction does not appear significantly in the full load of FIG.

Looking at the maximum strain distribution of FIGS. 2 and 3, it can be seen that deformation occurs locally at a specific portion of the rubber part, and the localization phenomenon of the strain has a problem of shortening the fatigue life of the conical rubber spring.

In addition, since the conical rubber spring used as the primary suspension device for the existing domestic railway vehicles is mainly imported from abroad, performance evaluation through product design and analysis for localization is required.

1. Republic of Korea Registered Patent Publication No. 10-1177337 (2012. 09. 07. announcement)

The present invention has been devised to solve the problems of the prior art as described above, and the object of the present invention is to optimize the curvature of each stage of the rubber part constituting the conical rubber spring to improve the load-deflection characteristics and rigidity, and reduce the strain. An object of the present invention is to provide an optimized design method of a conical rubber spring for a railway vehicle and a conical rubber spring designed thereby to increase the fatigue life.

Optimized design method of a conical rubber spring for a railway vehicle according to the present invention for achieving the above objects,

A modeling step of generating a 3D model of a conical rubber spring through reverse design using a 3D scan of a conical rubber spring for a railway vehicle, and a finite element model generation step of generating a finite element model for performing performance analysis of the modeled conical rubber spring. It is characterized by comprising an optimum design execution step to perform the optimal design of the conical rubber spring using the experimental design method.

At this time, the optimal design performing step is an experiment of constructing a conditional setting step for setting conditions necessary for performing optimal design and performance analysis, and constructing a three-level orthogonal array table in consideration of the number and level of design variables and the amount of experiment. Consisting of a planning step and an analysis and verification step to verify the optimal design by analyzing the performance analysis step and the performance analysis result to perform the performance analysis of the conical rubber spring according to the experimental plan according to the orthogonal array table. It is characterized by.

Here, the condition setting step, the target function setting step for setting the strain in the vertical, longitudinal and lateral stiffness analysis as a target function, and the first and second rubber layers constituting the rubber portion of the conical rubber spring Design parameters setting step of setting the total of eight curve radii on both sides of the curve formed at the lower end as design variables and setting constraints in the performance analysis result and the constraints in the modeling step performed in the performance analysis step Characterized in that it comprises a constraint setting step.

In addition, the level of each of the eight curve radius parameters set as the design variable in the design parameter setting step is characterized in that it is set to -10% and -20% based on the curve radius of the conventional conical rubber spring.

In addition, the constraints on the performance analysis result set in the step of setting the constraints are set such that the deflection amount of the conical rubber spring and the stiffness in the vertical, longitudinal and transverse directions are located within the standard deflection amount and stiffness range of the conical rubber spring. , The constraint in the modeling step is characterized by setting the inner and outer radii of the conical rubber spring and fixing the dimensions of the steel part including the shaft and the outer cylinder.

In addition, the performance analysis step is characterized by performing vertical stiffness, longitudinal stiffness and lateral stiffness analysis of the conical rubber spring under the tolerance and full load conditions.

And, in the analysis and verification step, performing an average analysis to analyze the effect of each design variable on the objective function, and performing variance analysis to verify the significance of the average analysis to verify the reliability of the optimal design results It is characterized by.

On the other hand, the conical rubber spring for a railway vehicle according to the present invention, in a conical rubber spring for a railway vehicle comprising a steel portion and a rubber portion composed of a first to third rubber layer, on the upper longitudinal section of the first rubber layer and the second rubber layer. The inner and outer curve radii of the curve to be formed are called r ₁ , r ₂ , r ₃ , and r ₄ , respectively, in order, and the inner and outer curve radii of the curve formed on the lower longitudinal cross sections of the first and second rubber layers, respectively. When r ₈ , r ₇ , r ₆ , and r ₅ , 10.45 ≤ r ₁ ≤ 10.72, 22.68 ≤ r ₂ ≤ 23.25, 22.84 ≤ r ₃ ≤ 23.35, 21.97 ≤ r ₄ ≤ 22.41, 10.20 ≤ r ₅ ≤ 10.43, 20.09 ≤ r ₆ ≤ 20.50, 15.33 ≤ r ₇ ≤ 15.72 and 13.84 ≤ r ₈ ≤ 14.11 (unit: mm).

According to the present invention, the curvature of each stage of the rubber part constituting the conical rubber spring is optimized to improve load-deflection characteristics and stiffness, and reduce strain to increase fatigue life.

In addition, according to the present invention, through the optimal design of the conical rubber spring, it is easy to manufacture and has the effect of reducing production cost.

1 is a perspective view showing a conventional conical rubber spring.
2 and 3 (a) ~ (c) is a view showing the vertical stiffness, longitudinal stiffness and lateral stiffness logarithmic strain distribution for the tolerance and full load conditions of the conventional conical rubber spring.
4 is a flow chart showing an optimized design method of a conical rubber spring for a railway vehicle according to the present invention.
5A to 5B are perspective and front cross-sectional views showing a conical rubber spring modeled in the modeling step of the present invention shown in FIG. 4.
6 (a) to 6 (b) are perspective and front sectional views showing a finite element model of a conical rubber spring generated in the finite element model generation step of the present invention shown in FIG. 4;
7 (a) and 7 (b) are views showing a comparison between a conventional conical rubber spring and an optimized conical rubber spring.
8 to 10 is a view showing a comparison of the strain distribution in the analysis of vertical stiffness, longitudinal stiffness and lateral stiffness for the tolerance load conditions of the conventional conical rubber spring and the optimized conical rubber spring.
11 to 13 are views showing a comparison of strain distribution in the analysis of vertical stiffness, longitudinal stiffness and lateral stiffness for full load conditions of a conventional conical rubber spring and an optimized conical rubber spring.
14 to 17 is a graph showing the average analysis results for the amount of deflection, vertical stiffness, longitudinal stiffness and transverse stiffness performed in the analysis and verification step of the present invention shown in FIG.
18 to 20 is a graph showing the average analysis results for the maximum strain in the analysis of vertical stiffness, longitudinal stiffness and lateral stiffness performed in the analysis and verification step of the present invention shown in FIG.

Hereinafter, with reference to the accompanying drawings will be described in detail the preferred design method of the conical rubber spring for the railway vehicle according to the present invention and preferred conical rubber spring designed thereby.

4 is a flow chart showing an optimized design method of a conical rubber spring for a railway vehicle according to the present invention, and FIGS. 5A to 5B are perspective views showing a conical rubber spring modeled in the modeling step of the present invention shown in FIG. 4. And a front sectional view, FIGS. 6 (a) to 6 (b) are perspective and front sectional views showing a finite element model of a conical rubber spring generated in the finite element model generation step of the present invention shown in FIG. a), (b) is a view showing a comparison between a conventional conical rubber spring and an optimized conical rubber spring, and FIGS. 8 to 10 are vertical stiffness for the tolerance load condition of the conventional conical rubber spring and the optimized conical rubber spring. , Comparison of strain distribution in longitudinal and transverse stiffness analysis, and FIGS. 11 to 13 are full load jaws of a conventional conical rubber spring and an optimized conical rubber spring. It is a diagram showing a comparison of strain distribution in the analysis of vertical stiffness, longitudinal stiffness and lateral stiffness for the gun, and FIGS. 14 to 17 are deflections, vertical stiffness, longitudinal stiffness and A graph showing the average analysis results for lateral stiffness, and FIGS. 18 to 20 show the average analysis results for the maximum strain in the analysis of vertical stiffness, longitudinal stiffness, and lateral stiffness performed in the analysis and verification steps of the present invention shown in FIG. 4. It is the graph shown.

The present invention is to optimize the curvature of each stage of the conical rubber spring used as a suspension device of a railway vehicle to reduce the localization of strain and to optimize the design method of a conical rubber spring for a railway vehicle to improve load-deflection characteristics and rigidity. And it relates to a conical rubber spring designed thereby, first, the optimized design method of the conical rubber spring for the railway vehicle according to the present invention, as shown in Figure 4, largely modeling step (S10), finite element model generation step (S20) and optimal It comprises a design execution step (S30).

In more detail, the modeling step (S10) relates to a step of modeling the conical rubber spring 100 for optimization, after performing a 3D scan on the conical rubber spring 100 to be optimized, using 3D scan Create 3D CAD models through reverse engineering.

At this time, in the modeling step (S10), after the 3D scan is performed before modeling the conical rubber spring 100, that is, the conventional conical rubber spring 100 is modeled as an initial model, in the experimental planning step (S34) to be described later. Various models are modeled according to the combination of design variables by the orthogonal array table.

5 (a), (b) is a perspective view and a front cross-sectional view of the initial model of the conical rubber spring 100 modeled in the modeling step (S10), respectively. It comprises 110 and a rubber part 120 made of rubber. The steel part 110 constitutes a central portion of the conical rubber spring 100 and is a hollow shaft 112 coupled to an axle box of a railway vehicle. ) And the reinforcement ring 116 inserted between the first and third rubber layers 122, 124, and 126 constituting the outer portion 114 and the rubber portion 120 coupled to surround the outer side of the rubber portion 120. It includes.

In addition, the rubber part 120 is coupled to the upper outer circumferential surface of the shaft 112 to serve as a spring, and is composed of a first rubber layer 122, a second rubber layer 124, and a third rubber layer 126 from the inside. , Each rubber layer (122, 124, 126) is formed so that the upper and lower ends of the longitudinal cross-section form a curved shape that curves inward, so that the entire rubber portion 120 forms a corrugated body.

Next, the finite element model generation step (S20) is a step of generating a finite element model of the conical rubber spring 100 modeled in the modeling step (S10) to perform performance analysis in the optimal design execution step (S30), which will be described later. In the present invention, a finite element model of a conical rubber spring 100 was created using Hypermesh 13.0 of Altair, a commercial pre-processing software.

When creating a finite element model, the steel part 110 created an element network using an 8-node solid element C3D8, and the rubber part 120 created an element network with a Hyprid cube solid element (C3D8H) for superelastic analysis. .

The size of the elements is 5mm on average, consisting of at least 0.3 to 7.5mm depending on the geometry, and the strain energy density function of the rubber part 120 is the polynomial quadratic that best fits the results of uniaxial and equiaxial biaxial tensile tests. Used.

6 (a) to 6 (b) show perspective and sectional front views of a finite element model of the conical rubber spring 100 generated in the finite element model generation step (S20) of the present invention shown in FIG. Model information and physical properties are shown in Table 1 and Table 2 below.

Item Number of nodes Number of elements Rubber part 25,140 18,236 Steel part 25,440 17,640 Sum 50,580 35,876

Density (ton / mm ³ ) Elastic modulus (GPa) Poisson's ratio Rubber part 1.22E-09 Polynomial model
(test data) 0.475 Steel part 8E-09 206 0.3

Next, the optimal design performing step (S30) relates to a step of performing an optimal design of the conical rubber spring 100 using an experimental planning method, wherein the experimental planning method is a planning method for an experiment, that is, a problem to be solved. It means to plan how to conduct an experiment, how to take data, and how to obtain the maximum information from the minimum number of experiments by analyzing the data with any statistical method.

There are one-way, two-way, factor-placement, and some implementation methods in the above-described experimental planning method. Some of these methods do not analyze unnecessary interactions instead of reducing the number of experiments as little as possible by selecting only a part of the level combination of each factor. As an example, an experiment by an orthogonal arrangement table to be described later belongs to this.

The optimal design performing step (S30) includes a condition setting step (S32), an experimental planning step (S34), a performance analysis performing step (S36), and an analysis and verification step (S38). S32) relates to a step of setting conditions necessary for performing modeling and performance analysis for optimal design.

In more detail, the condition setting step (S32) comprises an objective function setting step (S32a), a design variable setting step (S32b), and a constraint setting step (S32c), wherein the objective function setting step (S32a) Is a step of setting an objective function that means a function of an independent variable to be maximum or minimum in an optimization technique, and the present invention reduces strain and improves localization of strain as described above. The purpose was to set the strain at the time of stiffness analysis in each direction, that is, in the vertical direction, the longitudinal direction, and the transverse direction.

Next, the design parameter setting step (S32b) relates to the optimization of the conical rubber spring 100, that is, the step of setting the design variable to optimize the objective function, in the case of the conical rubber spring 100, the rubber part ( Since the curve radius of 120), that is, the curve radius of the curves formed at the top and bottom of the longitudinal sections of the first to third rubber layers 122, 124 and 126, respectively, is known to affect the stiffness and strain in each direction, in the present invention, FIG. As shown in b), a total of eight curved radii on both sides of the curve formed on the upper and lower longitudinal sections of the first rubber layer 122 and the second rubber layer 124, except for the third rubber layer 126, which has relatively little influence, are designed. It was set as a variable.

At this time, it is desirable to set the level of each factor set as a design variable to -10% and -20% based on the curve radius of the conventional conical rubber spring 100, that is, the initial model, because the factor is generally The level range of these is set to a value of ±, but if the radius of curvature of the rubber part 120 is increased, there is a possibility that the rubber does not sufficiently penetrate during injection molding after the mold is manufactured. It is set to the range of.

Next, the constraint setting step (S32c) relates to the step of setting a constraint in the objective function, that is, the performance analysis result performed in the performance analysis step (S36) and the modeling step (S10).

First, the characteristics to be confirmed when evaluating the performance of the conical rubber spring 100 in relation to the constraints on the performance analysis results are the amount of deflection and the stiffness in three directions, that is, the stiffness in the vertical, longitudinal, and transverse directions, so these characteristics are set as the objective function. However, when considering the test evaluation standards to be described later, the optimization of the conical rubber spring 100 does not maximize the stiffness, but may be located within the standard stiffness range, so it may be the primary purpose. It is desirable to set the stiffness of the direction as a limiting condition.

In addition, the limitation condition at the time of modeling is to fix the dimensions of the rubber portion 120, that is, the inner and outer radii of the rubber spring 100 and the dimensions of the steel portion 110 including the shaft 112 and the outer cylinder portion 114. It is preferable, because the size and shape of the bogie and axle box must also be modified when changing the part, so that the inner and outer radii of the rubber part 120 and the steel part 110 are saved for cost and time savings. Fixing the dimensions was set as a limiting condition in modeling and applied to the above-described modeling step (S10).

The conditions set in the condition setting step (S32) are summarized as follows.

, ... (1)

, ... (One)

... (2)

... (2)

... (3)

... (3)

... (4)

... (4)

는 각 강성 해석시의 변형률을 나타내는 것으로, 상기 (1)식은 조건 설정단계(S32)에서 설정된 목적함수와 설계변수를 나타낸 것이다.Here, r ₁ to r ₈ represent eight curve radii set as design variables,

Denotes the strain at the time of each stiffness analysis. The equation (1) shows the objective function and design variables set in the condition setting step (S32).

는 각 방향의 강성을 나타내는 것이고, L과 U는 각각 표준 강성 범위의 최소값과 최대값을 나타내며,

은 각각 고무부(120)의 내,외측 반경 및 스틸부(110)의 치수를 나타낸 것이다.In addition, the formulas (2) to (4) represent the set constraints, respectively.

Denotes the stiffness in each direction, L and U denote the minimum and maximum values of the standard stiffness range, respectively.

Indicates the inner and outer radii of the rubber part 120 and the dimensions of the steel part 110, respectively.

Next, the experiment planning step (S34) relates to a step of establishing an experiment, that is, planning for performance analysis, in consideration of the objective function, design variables, and constraints set in the condition setting step (S32). Considering the level (range) and the amount of experiments, the 3-level table of orthogonal arrays is constructed.

Orthogonal arrangement table of the three-level system used in the experiment plan in the present invention is generally represented by the following equation,

... (5)

... (5)

Here, m is an integer of 2 or more, 3 ^m represents the size of the experiment, and (3 ^m -1) / 2 represents the number of columns in the orthogonal array table.

In the present invention, as shown in Table 3 below, using the commercial statistical program Minitab 14, L ₂₇ orthogonal arrangement table was constructed.

Experiment
number Column number r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ One One One One One One One One One 2 One One One One 2 2 2 2 3 One One One One 3 3 3 3 4 One 2 2 2 One One One 2 5 One 2 2 2 2 2 2 3 6 One 2 2 2 3 3 3 One 7 One 3 3 3 One One One 3 8 One 3 3 3 2 2 2 One 9 One 3 3 3 3 3 3 2 10 2 One 2 3 One 2 3 One 11 2 One 2 3 2 3 One 2 12 2 One 2 3 3 One 2 3 13 2 2 3 One One 2 3 2 14 2 2 3 One 2 3 One 3 15 2 2 3 One 3 One 2 One 16 2 3 One 2 One 2 3 3 17 2 3 One 2 2 3 One One 18 2 3 One 2 3 One 2 2 19 3 One 3 2 One 3 2 One 20 3 One 3 2 2 One 3 2 21 3 One 3 2 3 2 One 3 22 3 2 One 3 One 3 2 2 23 3 2 One 3 2 One 3 3 24 3 2 One 3 3 2 One One 25 3 3 2 One One 3 2 3 26 3 3 2 One 2 One 3 One 27 3 3 2 One 3 2 One 2

Here, 1 is a conventional conical rubber spring, that is, the same value as the curve radius of the initial model, 2 means a value reduced by 10% from the curve radius of the initial model, and 3 is 20% from the curve radius of the initial model. It means reduced value.

Next, the performance analysis performing step (S36) relates to a step of performing a performance analysis for the optimization of the conical rubber spring 100 according to the experimental plan by the orthogonal array table established in the experimental planning step (S34). The analysis of vertical stiffness, longitudinal stiffness and lateral stiffness of the spring 100 was performed 27 times under tolerance load conditions and full load conditions, respectively, and a total of 162 performance analyzes were performed.

In addition, the analysis result for the strain was confirmed by the von-Mises logarithmic strain, and the performance analysis results are shown in Table 4 below.

Experiment
number Sag
(mm) tolerance Full car Strain (full scale) ranking Perpendicular
Rigid Bell
Rigid Transverse
Rigid Perpendicular
Rigid Bell
Rigid Transverse
Rigid Vertical stiffness analysis Longitudinal stiffness analysis Lateral stiffness analysis One 13.93 0.696 2.949 2.286 0.706 3.433 2.660 0.484 0.620 0.606 4 2 13.55 0.716 3.002 2.309 0.729 3.483 2.681 0.458 0.605 0.611 3 3 13.26 0.729 3.045 2.326 0.746 3.522 2.692 0.454 0.591 0.614 - 4 13.42 0.724 2.998 2.325 0.735 3.483 2.699 0.453 0.624 0.604 - 5 13.02 0.743 3.064 2.353 0.760 3.554 2.724 0.452 0.583 0.609 - 6 13.17 0.737 3.124 2.385 0.751 3.623 2.762 0.490 0.599 0.609 - 7 13.01 0.746 3.017 2.340 0.760 3.488 2.708 0.453 0.590 0.603 - 8 13.08 0.740 3.103 2.387 0.754 3.599 2.765 0.476 0.551 0.605 - 9 12.76 0.760 3.164 2.414 0.775 3.644 2.785 0.452 0.554 0.608 - 10 13.42 0.724 3.060 2.351 0.735 3.548 2.722 0.570 0.621 0.607 - 11 13.06 0.740 3.078 2.356 0.757 3.552 2.715 0.539 0.595 0.611 - 12 13.34 0.729 3.026 2.337 0.737 3.475 2.698 0.454 0.545 0.604 - 13 13.25 0.732 3.094 2.376 0.760 3.594 2.750 0.453 0.597 0.606 - 14 12.99 0.746 3.106 2.379 0.760 3.591 2.746 0.452 0.586 0.608 - 15 13.63 0.711 3.071 2.374 0.724 3.544 2.749 0.593 0.649 0.602 5 16 12.92 0.751 3.110 2.388 0.766 3.596 2.755 0.453 0.566 0.606 - 17 13.00 0.746 3.150 2.410 0.763 3.643 2.782 0.573 0.651 0.605 - 18 13.28 0.732 3.084 2.383 0.743 3.561 2.755 0.453 0.567 0.601 - 19 13.44 0.721 3.082 2.367 0.735 3.562 2.732 0.475 0.623 0.608 - 20 13.57 0.716 3.033 2.346 0.729 3.510 2.712 0.454 0.593 0.603 2 21 13.24 0.732 3.067 2.363 0.746 3.527 2.721 0.452 0.603 0.605 - 22 12.91 0.751 3.139 2.399 0.766 3.618 2.765 0.452 0.553 0.608 - 23 13.13 0.740 3.091 2.383 0.751 3.550 2.747 0.453 0.580 0.603 - 24 12.89 0.751 3.144 2.409 0.769 3.630 2.774 0.452 0.541 0.607 - 25 12.96 0.749 3.150 2.412 0.760 3.648 2.789 0.450 0.560 0.588 - 26 13.58 0.716 3.114 2.407 0.724 3.612 2.786 0.457 0.579 0.597 One 27 13.18 0.737 3.111 2.392 0.749 3.601 2.764 0.452 0.560 0.604 -

Meanwhile, in the performance analysis step (S36), performance analysis is performed according to the plan established in the experiment planning step (S34) in consideration of the conditions set in the above-described condition setting step (S32). The curve radius of the initial model, which is the standard for ₁ to r ₈ (corresponding to 1 in Table 3), and -10%, -20% of the curve radius (corresponding to 2 and 3 in Table 3, respectively) Is shown in Table 5 below.

Curve radius r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (early model) 13.229 28.705 25.658 22.193 11.460 20.295 19.400 13.975 2 (-10%) 11.90 25.83 23.09 19.97 10.31 18.27 17.46 12.58 3 (-20%) 10.58 22.96 20.53 17.75 9.17 16.24 15.52 11.18

In addition, the tolerance load condition and the full load condition in the performance analysis step (S36) are 15.68 kN and 25.48 kN, respectively. The same load as the load was applied, and in the lateral stiffness analysis, a vertical load was also applied, and a load corresponding to about 50% of the vertical load was applied in the lateral direction.

And, the standard amount of deflection of the conical rubber spring 100 and three directions, that is, vertical stiffness, longitudinal stiffness, and lateral stiffness, which are conditions for the performance analysis results, are shown in Table 6 below.

Item Reference value (limit conditions) Sag Tolerance and full load 15.0 ± 1.5mm Vertical stiffness Tolerance load 0.67 ± 0.10kN / mm Full load 0.64 ± 0.10kN / mm Species stiffness Tolerance load 3.0 ± 0.45kN / mm Full load 3.8 ± 0.57kN / mm Transverse stiffness Tolerance load 2.3 ± 0.35kN / mm Full load 2.8 ± 0.42kN / mm

As the analysis solver in the performance analysis step (S36), Dassault system's finite element analysis software Abaqus 6.12-1 was used, and although not shown, it imparts constraints such as a conical rubber spring 100 characteristic tester. For this purpose, all the degrees of freedom of the lower protruding portion of the conical rubber spring 100 were constrained, and the load was constant by connecting all nodes of the upper portion of the conical rubber spring 100 to the reference point with a rigid body-tie. Boundary conditions were constructed so that they could be delivered.

On the other hand, as can be seen in Table 4, except for Experiment No. 1, 2, 15, 20, and 26 experiments, all were found to violate the stiffness constraints shown in Table 6.

As shown in Table 7 below, the stiffness constraint is satisfied in experiment 15 of 5 experiments that satisfy the constraints including Experiment No. 1 corresponding to the initial model, but the strain is up to 22.6 when analyzing vertical and longitudinal stiffness. % Increase, the largest strain reduction in the 26 experiments, and the maximum 6.6% reduction in longitudinal stiffness analysis, the model No. 26 was selected as the optimal model.

Experiment
number Strain reduction Vertical stiffness analysis (%) Longitudinal stiffness analysis (%) Lateral stiffness analysis (%) One 0.0 0.0 0.0 2 5.3 2.5 -0.8 15 -22.6 -4.7 0.7 20 6.1 4.4 0.5 26 5.6 6.6 1.4

Optimal model r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 26 10.45 ~ 10.72 22.68 ~ 23.25 22.84 ~ 23.35 21.97 ~ 22.41 10.20 ~ 10.43 20.09 ~ 20.50 15.33 ~ 15.72 13.84 ~ 14.11

Table 8 above shows the design variables of Experiment No. 26, that is, values of r ₁ to r ₈ , respectively, selected as optimization models with reference to Tables 3 and 5, and r ₄ set to 1 in the orthogonal array table of Table 3 , r ₆ , r ₈ applied the same values as the initial model, and in the case of r ₃ and r ₅ set to 2 in the orthogonal array, a value reduced by 10% from the initial model was applied, and set to 3 in the orthogonal array. In the case of r ₁ , r ₂ , and r ₇ , values reduced by 20% from the initial model were applied.

In addition, a margin of ± 0.1% was applied in consideration of errors such as design errors and molding errors.

7 (a) and 7 (b) are shown by comparing the initial model and the optimized model, and it can be seen that many differences occurred in the parts corresponding to r ₁ to r ₈ , particularly r ₁ to r ₄ .

8 to 10 and FIGS. 11 to 13 are vertical stiffness and vertical stiffness for a conventional conical rubber spring 100, that is, an initial model and an optimized conical rubber spring 100 under tolerance load conditions and full load conditions, respectively. This diagram shows the comparison of strain distribution in the analysis of stiffness and lateral stiffness. It shows that the strain of the optimization model is reduced overall compared to the initial model in both tolerance and full load conditions, and that the distribution of strain that appeared locally after optimization is wider. Can be confirmed.

Next, the analysis and verification step (S38) relates to the step of verifying the reliability of the optimum design by analyzing the performance analysis results performed in the performance analysis step (S36).

In more detail, first, an analysis of means was performed to check the effect of each design variable, that is, r ₁ to r ₈ on the objective function, that is, the strain, and FIGS. 14 to 17 show deflection and vertical stiffness, respectively. , A graph showing the average analysis results for longitudinal stiffness and lateral stiffness, and FIGS. 18 to 20 are graphs showing average analysis results for the maximum strain in the analysis of vertical stiffness, longitudinal stiffness and lateral stiffness, respectively.

In addition, in order to verify the significance of the average analysis, analysis of variance was performed, and the results are shown in Tables 9 to 15 below.

Average analysis and variance table for deflection Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 13.24 13.42 13.21 13.37 13.25 13.43 13.19 13.35 2 (-10%) 13.21 13.16 13.24 13.23 13.22 13.17 13.25 13.22 3 (-20%) 13.21 13.09 13.22 13.07 13.19 13.06 13.23 13.1 delta 0.03 0.34 0.03 0.3 0.06 0.37 0.05 0.25 ranking 7 2 8 3 5 One 6 4 Dispersion
analysis p-Value 0.404 0.000 0.545 0.000 0.175 0.000 0.184 0.000

Average and variance analysis table for vertical stiffness Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 0.75 0.74 0.75 0.74 0.75 0.73 0.75 0.74 2 (-10%) 0.75 0.75 0.75 0.75 0.75 0.75 0.75 0.75 3 (-20%) 0.75 0.75 0.75 0.76 0.75 0.76 0.75 0.75 delta 0.00 0.02 0.00 0.02 0.00 0.02 0.00 0.01 ranking 7 2 6 3 8 One 5 4 Dispersion
analysis p-Value 0.482 0.000 0.298 0.000 0.751 0.000 0.288 0.001

Average and variance analysis table for species stiffness Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 3.54 3.51 3.56 3.56 3.55 3.52 3.55 3.58 2 (-10%) 3.57 3.58 3.57 3.56 3.57 3.57 3.56 3.56 3 (-20%) 3.58 3.60 3.56 3.57 3.57 3.60 3.58 3.55 delta 0.05 0.09 0.01 0.01 0.02 0.08 0.03 0.03 ranking 3 One 8 7 6 2 4 5 Dispersion
analysis p-Value 0.000 0.000 0.536 0.382 0.032 0.000 0.002 0.003

Average Analysis and Variance Analysis Table for Lateral Stiffness Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 2.72 2.70 2.74 2.74 2.73 2.72 2.73 2.75 2 (-10%) 2.74 2.75 2.74 2.74 2.74 2.74 2.74 2.74 3 (-20%) 2.75 2.77 2.74 2.74 2.74 2.75 2.75 2.73 delta 0.04 0.06 0.01 0.01 0.01 0.03 0.02 0.02 ranking 2 One 8 7 6 3 5 4 Dispersion
analysis p-Value 0.000 0.000 0.208 0.198 0.011 0.000 0.004 0.002

Average analysis and variance analysis table for maximum strain when analyzing vertical stiffness Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 0.465 0.472 0.472 0.474 0.471 0.473 0.467 0.508 2 (-10%) 0.504 0.472 0.480 0.461 0.479 0.457 0.474 0.463 3 (-20%) 0.443 0.469 0.461 0.478 0.462 0.484 0.473 0.442 delta 0.061 0.003 0.018 0.017 0.017 0.027 0.007 0.065 ranking 2 8 4 5 6 3 7 One Dispersion
analysis p-Value 0.050 0.753 0.867 0.946 0.893 0.768 0.899 0.029

Average analysis and variance analysis table for maximum strain when analyzing longitudinal stiffness Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 0.591 0.600 0.586 0.594 0.595 0.594 0.597 0.604 2 (-10%) 0.598 0.590 0.585 0.601 0.591 0.581 0.582 0.583 3 (-20%) 0.577 0.576 0.594 0.570 0.579 0.590 0.587 0.578 delta 0.021 0.024 0.009 0.031 0.016 0.013 0.015 0.025 ranking 4 3 8 One 5 7 6 2 Dispersion
analysis p-Value 0.133 0.092 0.214 0.001 0.089 0.247 0.266 0.682

Average analysis and variance analysis table for maximum strain in transverse stiffness analysis Average
analysis level r ₁ r ₂ r ₃ r ₄ r ₅ r ₆ r ₇ r ₈ 1 (standard) 0.608 0.608 0.607 0.604 0.604 0.603 0.606 0.605 2 (-10%) 0.606 0.606 0.604 0.606 0.605 0.607 0.604 0.606 3 (-20%) 0.603 0.602 0.605 0.606 0.605 0.607 0.606 0.605 delta 0.005 0.006 0.003 0.002 0.002 0.004 0.002 0.002 ranking 2 One 4 7 5 3 6 8 Dispersion
analysis p-Value 0.086 0.036 0.302 0.146 0.014 0.018 0.031 0.127

As for the amount of deflection in Table 9, it was found that r _6- > r _2- > r _4- > r ₈ had a significant effect, and only these variables showed significant p-values of 0.05 or less.

In addition, for the vertical stiffness in Table 10, as in the amount of deflection, r _6- > r _2- > r _4- > r ₈ had a significant effect on the stiffness, and only these variables have p-values of 0.05 or less, which is significant. Was confirmed.

The factor having the greatest influence on the longitudinal stiffness of Table 11 was r ₂ , and among the factors derived with p-values of 0.05 or less, r _2- > r _1- > r _6- > r _8- > r _7- > It was confirmed that there was an influence in the order of r ₅ , and the factor that had the greatest influence on the lateral stiffness in Table 12 was r ₂ as the longitudinal stiffness, and r ₃ and r4 had a p-value of 0.05 or more, indicating no significance. Among the significant factors, it was confirmed that there was an effect in the order of r _2- > r _6- > r _1- > r _7- > r _8- > r ₅ .

It was found that only r ₁ and r ₈ had a significant effect on the maximum strain in the vertical stiffness analysis shown in Table 13, and r ₄ had the largest effect on the maximum strain when analyzing the longitudinal stiffness shown in Table 14. , Was confirmed to be the only significant factor.

In addition, when analyzing the lateral stiffness shown in Table 15, the maximum strain was affected in the order of r _2- > r _1- > r _6- > r _3- > r _5- > r _7- > r _4- > r ₈ The higher ranks of r ₂ and r ₃ were found to be insignificant because the p-value was less than 0.05. This statistically means that r ₂ and r ₃ do not significantly affect the maximum strain.

Summarizing the above analysis results, it can be summarized that design variables excluding r ₃ affect analysis results such as deflection, stiffness, and strain. Among them, r ₂ represents the maximum strain during vertical stiffness and longitudinal stiffness analysis. Except, it can be seen that all are located within the second rank and are the most influential factors, and r ₄ also affects various performances.

However, since all of the design variables except r ₃ have an effect on the strain in the 3-way stiffness analysis, which is an objective function, and performance evaluation items are very diverse, it is determined whether the effect of the characteristic design variables on the overall performance evaluation is large or small. I can't.

Therefore, it can be concluded that the combination of various design variables has an effect on performance improvement, and the optimal design result with model 26 selected as the optimal model can be trusted.

According to the optimization design method of the conical rubber spring for railway vehicles according to the present invention and the conical rubber spring 100 designed thereby, the curvature of each stage of the rubber part 120 constituting the conical rubber spring 100 is optimized. By improving the load-sag characteristics and stiffness, and reducing the strain rate to increase the fatigue life, it is also easy to manufacture and reduce the production cost through the optimal design of the conical rubber spring, and mainly from overseas It has various advantages such as contributing to the localization development of the conical rubber spring 100 that relies on import.

The above-described embodiments have been described with respect to the most preferred examples of the present invention, but are not limited to the above-described embodiments, and do not deviate from the technical spirit of the present invention, such as programming the optimization design method according to the present invention and storing it in a recording medium. It is apparent to those skilled in the art that various modifications are possible within the scope.

The present invention relates to an optimized design method of a conical rubber spring for a railway vehicle and a conical rubber spring designed thereby, more specifically, to optimize the curvature of each stage of a conical rubber spring used as a suspension device for a railway vehicle to localize the strain. The present invention relates to an optimized design method of a conical rubber spring for a railway vehicle and a conical rubber spring designed thereby to reduce and improve load-sag characteristics and rigidity.

100: conical rubber spring 110: steel
112: shaft 114: outer cylinder
116: reinforcement ring 120: rubber part
122: first rubber layer 124: second rubber layer
126: third rubber layer S10: modeling step
S20: Finite element model creation phase S30: Optimal design execution phase
S32: Condition setting step S32a: Objective function setting step
S32b: Design variable setting step S32c: Constraint setting step
S34: Experiment planning step S36: Performance analysis execution step
S38: Analysis and verification stage

Claims (8)

A modeling step of generating a 3D model of a conical rubber spring through reverse design using a 3D scan of the conical rubber spring for a railway vehicle,
A finite element model generation step for generating a finite element model for performance analysis of the modeled conical rubber spring,
An optimum design method of a conical rubber spring for a railway vehicle, characterized in that it comprises an optimum design execution step of performing an optimum design of a conical rubber spring using an experimental planning method.
According to claim 1,
The optimal design execution step,
A condition setting step for setting conditions necessary for performing optimal design and performance analysis;
An experiment planning step of constructing a three-level orthogonal array table considering the number and level of design variables and the amount of experiments;
Performance analysis step of performing the performance analysis of the conical rubber spring according to the experimental plan according to the orthogonal array table and
Optimized design method of a conical rubber spring for a railway vehicle, characterized by comprising an analysis and verification step to verify the optimum design by analyzing the performance analysis results.
According to claim 2,
The condition setting step,
The objective function setting step of setting the strain in the vertical, longitudinal and lateral stiffness analysis as the objective function,
Design parameter setting step of setting a total of eight curve radii on both sides of the curve formed at the upper and lower ends of the first and second rubber layers constituting the rubber portion of the conical rubber spring, and
Optimization design method of a conical rubber spring for a railway vehicle, characterized in that it comprises a constraint setting step of setting a constraint in the modeling step and a constraint condition for the performance analysis result performed in the performance analysis step.
According to claim 3,
The level of each of the eight curve radius parameters set as the design variable in the design parameter setting step is set to -10% and -20% based on the curve radius of the conventional conical rubber spring. Optimization design method.
According to claim 3,
The limiting condition for the performance analysis result set in the setting of the above-mentioned constraint conditions is set to be that the deflection amount of the conical rubber spring and the stiffness in the vertical, longitudinal and transverse directions are located within the standard deflection amount and stiffness range of the conical rubber spring,
The limiting condition in the modeling step is an optimization design method of a conical rubber spring for a railway vehicle, characterized in that the inner and outer radii of the conical rubber spring and the dimensions of the steel part including the shaft and the outer cylinder are fixed.
According to claim 2,
In the performance analysis step, an optimization design method of a conical rubber spring for a railway vehicle is characterized by performing vertical stiffness, longitudinal stiffness and lateral stiffness analysis of a conical rubber spring under tolerance and full load conditions.
According to claim 3,
In the analysis and verification step, it is characterized by performing an average analysis to analyze the effect of each design variable on the objective function, and performing a variance analysis to verify the significance of the average analysis to verify the reliability of the optimal design result. Optimized design method of conical rubber spring for railway vehicles.
In the conical rubber spring for a railway vehicle comprising a steel portion and a rubber portion consisting of a first to third rubber layer,
The inner and outer curve radii of the curves formed on the upper longitudinal sections of the first rubber layer and the second rubber layer are respectively referred to as r ₁ , r ₂ , r ₃ , and r ₄ , respectively.
When the inner and outer curve radii of the curve formed on the lower longitudinal cross-sections of the first rubber layer and the second rubber layer are r ₈ , r ₇ , r ₆ and r ₅ , respectively,
10.45 ≤ r ₁ ≤ 10.72, 22.68 ≤ r ₂ ≤ 23.25, 22.84 ≤ r ₃ ≤ 23.35, 21.97 ≤ r ₄ ≤ 22.41, 10.20 ≤ r ₅ ≤ 10.43, 20.09 ≤ r ₆ ≤ 20.50, 15.33 ≤ r ₇ ≤ 15.72 and 13.84 ≤ r ₈ ≤ 14.11 (unit: mm) Conical rubber spring for railway vehicles.

Images