The coupling of deformable structures is an old topic of discussion, but many difficulties are still encountered today. Working in the framework of linear vibrations, and after several simplifications, the problem is formulated by means of substructuring techniques which require only the dynamic behaviour of each uncoupled structure. In these conditions, it is possible, in theory, to predict the behaviour of the assembled system. Models to predict the forces acting on the boundary of a receiving structure coupled with a vibrating component contain matrices and vectors - the parameters - built from measurements, which themselves contain errors. These errors propagate through the model to the predicted forces. To ensure that the level of forces complies with specifications on their magnitude, we should know their highest value. But, in presence of the quoted errors and therefore in presence of dispersion in the forces, how can the most recurrent highest value be estimated. To answer this question, simulations provide a set of predicted forces, from which a scatter plot of their norm versus that of parameters is drawn. To effectively extract information from the scatter plot, a density function is defined to neglect points of rare occurrence. This approach is compared with a Monte Carlo method in which information is extracted from a histogram of the norm of predicted forces. This comparison is made in the case offerees acting on the chassis of a vehicle because of a running fan system engine.
Prediction of forces acting on the chassis of a vehicle: comparison of two methods for estimating the dispersion of the results
2009
7 Seiten, 5 Bilder, 7 Quellen
Conference paper
English
Chassis vehicle, chassis vehicle control system and chassis vehicle control method
European Patent Office | 2021
|