| HYBRID SHAPE OPTIMIZATION OF AUTOMOTIVE SPRING SEAT
|
| Przemysław Sebastjan , Wacław Kuś |
| Department for Computational Mechanics and Engineering RMT4, Silesian University of Technology |
|
|
|
|
|
|
|
| ABSTRACT |
|
Work presented in this article concerns an industrial example of a mass optimization problem, performed on
automotive suspension components. Due to the fact, that the considered element, i.e. spring seat is a part of the vehicle
suspension module, analyses are performed on the shock absorber assembly level with the mutual interaction of each body.
To solve constrained optimization task, a combination of genetic and gradient-based algorithms are used, involving nonlinear
static strength problem, represented by finite element method analysis. Parametric representation of considered component
geometry is used, incorporating manufacturing restrictions and design assembly requirements. The final, optimized form of
the spring seat is then compared with its initial shape, i.e. “working space”, especially under constraints fulfillment, which are
specified by the customer. Performance comparison of used methods (or combination of such) is the aim of this paper,
focusing on practical aspects of work, i.e. final mass decrease, number of function calls, or design space exploration
capabilities. The optimization method presented hereby is universal and can be adapted to many fields of industry, wherever
efficient mass reduction must be achieved incorporating the feasibility of the component. The hybrid optimization approach
presented in the paper allows effective mass reduction and works well for multimodal objective functions typically found in
industrial applications. |
| Key Words:
Shape optimization, Hybrid optimization, Genetic algorithms, Finite element method, Automotive suspension, Non-linear optimization
|
|
Corresponding Author.
Wacław Kuś , Email. 
Preview
Full text via DOI
Download Citation 
Print 
