Use of Genetic Algorithms for the Multicriteria Optimal Design of Multibody Systems

The design step of multibody systems requires
in some specific cases an optimization process, in order to determine the set of parameters which lead to optimal kinematic
or dynamic performances. The optimization task is complicated by the explicit time-dependent of the differential equations of motion, the constraints and/or the objective
functions. Usually the designer have to hoose one criterium to optimize; the others are considered as technologic constraints.
This method is well adapted to take into account the boundary constraints that put into practice the range of variation of some design parameters, such as stiffness and
damping of shock absorbers.
For constraints depending on the behaviour of the multibody system, such method means that the designer has to limit the behaviour a priori without knowing its effect on the optimal solution. This generally leads to an incomplete view of the feasible well adapted solutions. So it may be better to consider the behaviour-dependent constraints also as
the objective functions, such as the displacement, the velocity and the acceleration of the bodies during the time. Such approach leads to the multicriteria optimal design. Usually the obtained set of solutions is called non dominated. It gives
the designer the choice between several feasible possibilities that are not better than an another. The aim of this paper is to propose a multicriteria optimal design method adapted to general multibody systems and submitted to kinematic and/or dynamic time-dependent criteria. The optimization process is based on genetic algorithms.
The paper especially describes the way for taking into account the time-dependent characteristics of the criteria.
The first result refers to the well-known benchmark of the shock absorber. Comparison between the single and multicriteria solutions gives a good understanding of the advantages of the proposed approach. Illustrative examples are given in the context of the optimization of a motorcar
suspension and the lateral dynamics of an urban railway vehicle.