Model Reduction of Contact Dynamics Simulation Using Modified Lyapunov Balancing Method

Author(s):  
Jianxun Liang ◽  
Ou Ma

Finite element models can accurately simulate impact-contact dynamics response of a multibody dynamical system. However, such a simulation remains very inefficient because very small integration time step must be used when solving the involved differential equations. Although many model reduction techniques can be used to improve the efficiency of finite element based simulations, most of these techniques cannot be readily applied to contact dynamics simulations due to the high nonlinearity of the contact dynamics model. This paper presents a model reduction approach for finite-element based multibody contact dynamics simulation, based on a modified Lyapunov balanced truncation method. An example is presented to demonstrate that, by applying the model reduction the simulation process is significantly speeded up and the resulting error is bounded within an acceptable level. The performance of the method with respect to some influential factors such as element size, shape and contact stiffness is also investigated.

Author(s):  
Jianxun Liang ◽  
Ou Ma ◽  
Caishan Liu

Finite element methods are widely used for simulations of contact dynamics of flexible multibody systems. Such a simulation is computationally very inefficient because the system’s dimension is usually very large and the simulation time step has to be very small in order to ensure numerical stability. A potential solution to the problem is to apply a model reduction method in the simulation. Although many model reduction techniques have been developed, most of them cannot be readily applied due to the high nonlinearity of the involved contact dynamics model. This paper presents a solution to the problem. The approach is based on a modified Lyapunov balanced truncation method. A numerical example is presented to demonstrate that, by applying the proposed model reduction method, the simulation process can be significantly speeded up while the resulting error caused by the model reduction is still within an acceptable level.


2016 ◽  
Vol 28 (14) ◽  
pp. 1886-1904 ◽  
Author(s):  
Vijaya VN Sriram Malladi ◽  
Mohammad I Albakri ◽  
Serkan Gugercin ◽  
Pablo A Tarazaga

A finite element (FE) model simulates an unconstrained aluminum thin plate to which four macro-fiber composites are bonded. This plate model is experimentally validated for single and multiple inputs. While a single input excitation results in the frequency response functions and operational deflection shapes, two input excitations under prescribed conditions result in tailored traveling waves. The emphasis of this article is the application of projection-based model reduction techniques to scale-down the large-scale FE plate model. Four model reduction techniques are applied and their performances are studied. This article also discusses the stability issues associated with the rigid-body modes. Furthermore, the reduced-order models are utilized to simulate the steady-state frequency and time response of the plate. The results are in agreement with the experimental and the full-scale FE model results.


2019 ◽  
Vol 799 ◽  
pp. 211-216
Author(s):  
Alina Sivitski ◽  
Priit Põdra

Contact modeling could be widely used for different machine elements normal contact pressure calculations and wear simulations. However, classical contact models as for example Hertz contact models have many assumptions (contact bodies are elastic, the contact between bodies is ellipse-shaped, contact is frictionless and non-conforming). In conditions, when analytical calculations cannot be performed and experimental research is economically inexpedient, numerical methods have been applied for solving such engineering tasks. Contact stiffness parameters appear to be one of the most influential factors during finite element modeling of contact. Contact stiffness factors are usually selected according to finite element analysis software recommendations. More precise analysis of contact stiffness parameters is often required for finite element modeling of contact.


Author(s):  
Gyeongho Kim ◽  
DongOk Kim ◽  
Woo-Seok Choi ◽  
Ji Ho Kang ◽  
Jae Man Noh

Graphite blocks are important core components of the high temperature gas-cooled reactor. As these blocks are simply stacked in array, collisions among neighboring components may occur during earthquakes or accidents. Thus, it is important to develop a reliable seismic model of the stacked graphite blocks and have them designed to maintain their structural integrity during the anticipated occurrences. Various aspects involved in modeling and calculating impact-contact dynamics can affect the resulting behavior of the graphite block. These include mesh size, time step, contact behavior, mechanical constraint formulation of impact-contact analysis, etc. This work is dedicated to perform comparative studies and the effects of these parameters will be identified. The insights obtained through these studies will help build a realistic impact-contact model of the graphite block, from which a lumped or reduced dynamics model will be developed for the seismic analysis of the reactor including these graphite components.


2018 ◽  
Vol 10 (1) ◽  
pp. 168781401774807 ◽  
Author(s):  
Yonghu Wang ◽  
Dongwei Shu ◽  
Yusaku Fujii ◽  
Akihiro Takita ◽  
Tsuneaki Ishima ◽  
...  

In order to precisely measure water impact loads of a spherical structure vertically dropping onto a calm water surface, a new validity check of the analysis using the levitation mass method experiment is proposed. The main feature of levitation mass method experiment is to obtain a better estimation of early water impact loads through the application of Doppler effect. Experimental results of different heights are verified based on the Assessment Index and are in comparison with the classical experimental data for validation purpose. It shows that the levitation mass method measurement is useful and effective to obtain the water impact loads for the crashworthiness analysis. Besides, early water impact hydrodynamic behaviors are simulated based on the nonlinear explicit finite element method, together with application of a multi-material arbitrary Lagrangian–Eulerian solver. A penalty coupling algorithm is utilized to realize fluid–structure interaction between the spherical body and fluids. Convergence studies are performed to construct the proper finite element model by the comparison with experimental results, where mesh sensitivity, contact stiffness, and time-step size parametric studies are thoroughly investigated. The comparisons between experimental and numerical results show good consistency by the prediction of the water impact coefficients on the structure.


Sign in / Sign up

Export Citation Format

Share Document