Refine your search:     
Report No.
 - 
Search Results: Records 1-2 displayed on this page of 2
  • 1

Presentation/Publication Type

Initialising ...

Refine

Journal/Book Title

Initialising ...

Meeting title

Initialising ...

First Author

Initialising ...

Keyword

Initialising ...

Language

Initialising ...

Publication Year

Initialising ...

Held year of conference

Initialising ...

Save select records

Journal Articles

An Adaptive control system using the fuzzy theory for transient multi-physics numerical simulations

Muramatsu, Toshiharu; Yagawa, Genki*

International Journal for Numerical Methods in Fluids, 54(6-8), p.805 - 830, 2007/07

 Times Cited Count:0 Percentile:0.02(Computer Science, Interdisciplinary Applications)

An adaptive control system to yield optimum time step sizes was developed using the fuzzy theory for transient multi-physics numerical simulations. Applications of the control system reveals considerable amount of the computing time savings, typically by 50 % to 75 % of the computing time required when the time step size was not controlled by the system. The result obtained in this work is very encouraging in the sense that the adaptive control system would be used as one of the efficient measures for saving computing time when one wishes to perform extremely large-scale computations in transient multi-physics numerical simulations.

Oral presentation

Part/component-based large-scale finite element analysis; Discontinuous meshes stitching up

Tian, R.

no journal, , 

In many situations, using a non-matching mesh is convenient even indispensable. However, conventional finite element interpolation fails when meshes do not match and a gluing algorithm is required to enforce inter-mesh continuity. In this paper, a gluing algorithm is developed based on a meshless interpolation. A continuous function is constructed across two non-matching and discontinuous meshes by a meshless interpolation using nodes from both sides to perform the stitch-up. The original meshes are used for such purposes as the integration of weak form and construction of mass matrix in dynamic analyses etc. In this study, a compactly supported radial basis function interpolation is employed. The gluing algorithm developed is tested in static and wave propagation problems. Some numerical evaluation in three dimensions are also provided. Through the numerical samples, superior performance in both accuracy and convergence over traditional approaches are demonstrated. Compared with other gluing algorithms, for example, the Lagrange multipliers, the current algorithm offers (1) straightforward implementation in any dimensions, and (2) banded, positive and definite system matrices, posing no difficulty in equation solvers.

2 (Records 1-2 displayed on this page)
  • 1