Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/23278
Title: Data structures and program techniques of finite element methods for analysis and optimization of electric devices
Authors: Fu, W 
Chen, Y
Keywords: Date structure
Electromagnetic field
Finite element method
Mesh generation
Programming
Sub-matrix
Issue Date: 2015
Publisher: IOS Press
Source: International journal of applied electromagnetics and mechanics, 2015, v. 47, no. 4, p. 875-883 How to cite?
Journal: International journal of applied electromagnetics and mechanics 
Abstract: State-of-the-art programming techniques of finite element methods (FEM) for electromagnetic field computation are presented. It covers program structure, data structure and algebraic matrix equation solver. The advantages of the proposed program structure are that multi-developers are empowered to work on different solvers and share common algorithms. The beauty of the proposed data structure are that two dimensional (2-D) FEM, multi-slice FEM and three-dimensional (3-D) FEM can share the same data structure. It allows quick access to all data and is efficient for organizing FEM programs and convenient for mesh generation. It can also deal with motion problems readily. The merits of the matrix equation solver are that it can automatically deal with Dirichlet boundary conditions, master-salve boundary conditions and all other constraints in sparse matrix equations. The sub-matrix operation technique allows the electric circuit equations to be coupled easily with electromagnetic field equations. To couple the FEM with optimization method, a parameterized mesh technique is proposed to avoid mesh regeneration in the optimization progress.
URI: http://hdl.handle.net/10397/23278
ISSN: 1383-5416 (print)
DOI: 10.3233/JAE-130131
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

53
Last Week
2
Last month
Checked on Sep 17, 2017

Google ScholarTM

Check

Altmetric



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.