This contribution deals with improvements which have been undertaken on a modelling and simulation tool (developed by the authors), aimed to numerically characterize the dynamic behaviour of rubber to metal devices (such as: engine mountings, shock absorbers, vibration dampers, etc.). The adopted approach allows designing such devices in a CAD like environment, meshing them and simulating their dynamics; materials constitutive equations modelling is operated by a neural network, working directly on numerical data supplied by experimentation. The peculiar properties of rubber, in terms of non-linearity, viscoelasticity as well as frequency, temperature and static strain dependence, are taken into account in the numerical analysis. The proposed method avoids the implementation of finite elements models (as proposed by many authors) in order to produce a light and easy-to-use modelling and simulation tool, to be adopted by industrial technicians in the design optimization of the mentioned devices. To achieve this point, discrete models, constructed by a multi-body particle based approach, are applied. Several models of such a kind have been taken into account during this research project, from discrete concentrated parameters models to a chosen hybrid particle based model, where local finite elements stiffness is computed directly by the continuum constitutive equations. Simulative results are successfully compared with experimental data for some existing devices, provided by CF Gomma S.p.A (Brescia, ITALY).
Discrete models for the simulation of rubber components dynamics
VETTURI, David;MAGALINI, Andrea;LANCINI, Matteo
2006-01-01
Abstract
This contribution deals with improvements which have been undertaken on a modelling and simulation tool (developed by the authors), aimed to numerically characterize the dynamic behaviour of rubber to metal devices (such as: engine mountings, shock absorbers, vibration dampers, etc.). The adopted approach allows designing such devices in a CAD like environment, meshing them and simulating their dynamics; materials constitutive equations modelling is operated by a neural network, working directly on numerical data supplied by experimentation. The peculiar properties of rubber, in terms of non-linearity, viscoelasticity as well as frequency, temperature and static strain dependence, are taken into account in the numerical analysis. The proposed method avoids the implementation of finite elements models (as proposed by many authors) in order to produce a light and easy-to-use modelling and simulation tool, to be adopted by industrial technicians in the design optimization of the mentioned devices. To achieve this point, discrete models, constructed by a multi-body particle based approach, are applied. Several models of such a kind have been taken into account during this research project, from discrete concentrated parameters models to a chosen hybrid particle based model, where local finite elements stiffness is computed directly by the continuum constitutive equations. Simulative results are successfully compared with experimental data for some existing devices, provided by CF Gomma S.p.A (Brescia, ITALY).File | Dimensione | Formato | |
---|---|---|---|
FullPaper ECCM2006.pdf
gestori archivio
Tipologia:
Full Text
Licenza:
DRM non definito
Dimensione
640.02 kB
Formato
Adobe PDF
|
640.02 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.