The best solution to the axle contact problem of F

2022-07-27
  • Detail

Solving the axle contact problem of Ford Motor vehicle

in the design process of a new part, the precise boundary conditions are usually not given, which may lead to the problem of nonlinear contact conditions involving contact problems. In the iterative process of component optimization, the contact conditions and force transfer path will change with the continuous modification of geometric shape, so the influence of contact conditions must be considered

a simple method is to simplify the contact conditions into node forces, but this will lead to unsatisfactory optimization analysis results. Moreover, this is a time-consuming and labor-consuming process, which must be avoided in the rapid and efficient development process. Therefore, it is a general trend to add direct contact conditions in the optimization area of structural optimization design

The modeling method of Tosca enables us to deal with this problem in various finite element solvers as usual. Users first build an analysis model with the required contact conditions based on their own solver, and then define the optimization task. In the optimization algorithm, there is no need to deal with the existing contact conditions in the model. Through the contact force and its resultant stress, the contact problem is implicit in the optimization process. This is bound to give full play to the functional advantages of the contact algorithm used by Tosca users

tosca is a standard nonparametric optimization system, which can optimize the topology and shape of finite element models under arbitrary loads and constraints, and optimize the fringes of thin-walled structures. Tosca does not need to parameterize the model in the optimization process, which greatly reduces the workload and improves the adaptability of the optimal structure. The optimization algorithm based on the mechanical optimization standard makes the optimization process fast and stable

using Tosca for structural optimization design is an iterative process. In each iteration step, an external finite element solver is used to calculate the mechanical response of the structure. High quality calculation results are obtained by using standard solvers approved by the industry, including ABAQUS, ANSYS, I-DEAS, stran, Marc, etc. Another advantage of this subsequent consolidation is that users can work in their familiar solver and pre-processing environment without training to be familiar with another unfamiliar software environment. The existing finite element model can be directly applied to optimization calculation

through the interaction of various programs in Tosca, the closed-loop optimization design process from concept to finished product in the cad/cae system of new product structure can be completed. As shown in Figure 1

Figure 1 Tosca optimization design analysis flow chart

the introduction of structural optimization tools in the early design and development stage will give full play to the potential of CAE design analysis, and its topology and shape optimization technology will improve development efficiency and product performance. The product will be lighter, stronger and more stable.

by directly calling all the major finite element solvers today, Tosca optimization system provides a set of comprehensive solutions for solving the optimization problems of actual structures, and has been widely agreed. Through the interface with the nonlinear software ABAQUS, Tosca can easily solve the structural optimization problem with contact conditions

tosca can directly optimize the contact surface to make its contact force more uniform. Users can use ape module to directly optimize the contact surfaces of different shapes to eliminate the abnormal distribution of contact force. This technology can be used to optimize the design of such problems as interference fit and the connection of parts of different sizes

Ford Motor Company provides an ABAQUS model of shaft contact problem, which is a part of its 1.25/1.4 zetec-se engine, including 8 parts, which are connected with each other through contact definitions, as shown in Figure 2. The model is the final design scheme of the parts, so the stress level and other properties have been fully improved manually, and basically meet the design requirements

in this case, abaus6.1 is used, and a total of 4 hours are calculated on the HP c3700 workstation

the whole loading process is divided into the following 6 steps:

1. squeeze the piston pin, Tension bolt

2. determine the screw load

3. initialize the contact under tensile load by applying a control displacement

4. solve the mating force and inertia force under maximum acceleration (maximum tensile load)

5. initialize the contact under compression load by applying a control displacement

6. solve the air pressure and inertia force under maximum torque (maximum compression load)

it should be noted that, The inertia force truly simulates the real inertia load imposed on the unit. During the optimization process, the correction of the geometric shape will cause the actual stress state to be corrected timely and accurately

Figure 2 Ford 1.25/1.4 zetec-se engine parts shaft contact model

Figure 3 von Mises stress distribution in the initial design

Figure 4 von Mises stress distribution after five optimization design cycles

it should also be noted that the model used in the optimization process does not need any correction, which makes it very easy for ABAQUS users to start, and its various very complex models can be directly used for optimization analysis and design. The same is true for other finite element users

Tosca can replace the traditional manual method to modify and improve the original design of parts. Firstly, an existing finite element model is analyzed, and then based on the analysis results, it is continuously improved with reference to Tosca optimization standard until a satisfactory design scheme is obtained. This process is completed automatically, which is bound to greatly shorten the design and development cycle

the reason why ape shape optimization design is called "nonparametric" is that the entire optimization area is simply designated as multiple nodes whose positions can be modified, and there is no need to define variable specific geometric parameters or shape basic vectors. Such a parametric optimization method is bound to lead to restrictions on the designable space

the optimization objective of Ford 1.25/1.4 zetec-se engine parts is to minimize the maximum von Mises stress in steps 4 and 6 by modifying the inner contour of the connecting rod. This will enable the parts to provide sufficient strength margin for Ford to replace more powerful engines in the future

the optimization area is selected as the nodes on all inner contour lines defined by a node group. The element on the connecting rod is defined as a lattice adaptive smoothing region, so as to ensure that the node position of the region involved will still be a high-quality lattice partition system after being corrected. The whole optimization process is terminated after 5 iteration cycles

the maximum von Mises stress distribution before optimization is shown in Figure 3. The stress concentration caused by compression load can be seen on the inner contour line near the crankshaft. Figure 4 shows the maximum von Mises stress distribution for the modified design after five optimization iteration cycles. It is obvious that the stress distribution in the design area tends to be balanced and reasonable

the variation trend of the maximum von Mises stress under tensile and compressive loads during the optimization process is shown in Figure 5. It can be seen that the maximum stress is reduced to 83% of the original design scheme in only 5 design cycles. Using Tosca to optimize design is fast and efficient, which can greatly shorten the design and development time and improve product performance

it should be emphasized that, as mentioned above, the initial optimization model has been improved many times by traditional manual methods. Obviously, if the exact boundary conditions are not given in the design process of a new part at the beginning, there may be a nonlinear contact condition problem involving contact problems. In the iterative process of component optimization, the contact conditions and force transfer path will change with the continuous modification of geometric shape, so the influence of contact conditions must be considered

a simple method is to simplify the contact conditions into node forces, but this will lead to unsatisfactory optimization analysis results. Moreover, this is a time-consuming and labor-consuming process, which must be avoided in the rapid and efficient development process. Therefore, it is a general trend to add direct contact conditions in the optimization area of structural optimization design

The modeling method of Tosca enables us to deal with this problem in various finite element solvers as usual. Users first build an analysis model with the required contact conditions based on their own solver, and then define the optimization task. In the optimization algorithm, there is no need to deal with the existing contact conditions in the model. Through the contact force and its resultant stress, the contact problem is implicit in the optimization process. This is bound to give full play to the functional advantages of the contact algorithm used by Tosca users

tosca can directly optimize the contact surface to make its contact force more uniform. Users can use ape module to directly optimize the contact surfaces of different shapes to eliminate the abnormal distribution of contact force. This technology can be used to optimize the design of such problems as interference fit and the connection of parts of different sizes

maximum von Mises stress in the design area in steps 4 and 6 of Figure 5

Ford Motor Company provides an ABAQUS model of shaft contact problem, which is a part of its 1.25/1.4 zetec-se engine, including 8 parts, which are connected with each other through contact definitions, as shown in Figure 2. The model is the final design scheme of the parts, so the stress level and other properties have been fully improved manually, and basically meet the design requirements

fig.6 connecting rod designable space

Figure 7 topology optimization results (front view)

figure 8 topology optimization results (rear view)

Figure 9 topology optimization results with manufacturing process constraints

abaus6.1 version is adopted in this case, and a total of 4 hours are calculated on HP c3700 workstation

the whole loading process is divided into the following 6 steps:

1. squeeze the piston pin, Tension bolt

2. determine the screw load

3. initialize the contact under tensile load by applying a control displacement

4. solve the mating force and inertia force under maximum acceleration

(maximum tensile load)

5. initialize the contact under compressive load by applying a control displacement

6. solve the air pressure and inertia force under maximum torque (maximum compression load)

it should be noted that, The inertia force truly simulates the real inertia load imposed on the unit. During the optimization process, the correction of the geometric shape will cause the actual stress state to be corrected timely and accurately

it should also be noted that the model used in the optimization process does not need any modification, which makes it very easy for ABAQUS users to start, and its various very complex models can be directly used for optimization analysis and design. The same is true for other finite element users

Tosca can replace the traditional manual method to modify and improve the original design of parts. Firstly, an existing finite element model is analyzed, and then based on the analysis results, it is continuously improved with reference to Tosca optimization standard until a satisfactory design scheme is obtained. This process is completed automatically, which is bound to greatly shorten the design and development cycle

the reason why ape shape optimization design is called "non parameterization", even if the solution can be learned faster, is that the entire optimization area is simply designated as multiple nodes whose positions can be modified, and there is no need to define variable specific geometric parameters or shape basic vectors. Such parameterization

Copyright © 2011 JIN SHI