Scribd
Upload
81 views4 pages

4.1.3 Shape Function: 4.2 Discretization of The 4.1 The Finite Element 4.1.2 Assembly

The document discusses shape functions in finite element analysis. Shape functions are used to interpolate values between mesh nodes. Linear shape functions are commonly used. For 3D simulations using tetrahedral meshes, linear shape functions must be defined for each tetrahedron. A coordinate transformation simplifies calculations by producing fixed, known nodal shape functions and requiring only calculation of the Jacobian matrix between coordinate systems.

Uploaded by

Paulo Henrique Mareze
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
81 views4 pages

4.1.3 Shape Function: 4.2 Discretization of The 4.1 The Finite Element 4.1.2 Assembly

The document discusses shape functions in finite element analysis. Shape functions are used to interpolate values between mesh nodes. Linear shape functions are commonly used. For 3D simulations using tetrahedral meshes, linear shape functions must be defined for each tetrahedron. A coordinate transformation simplifies calculations by producing fixed, known nodal shape functions and requiring only calculation of the Jacobian matrix between coordinate systems.

Uploaded by

Paulo Henrique Mareze
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

 

     
Next: 4.2 Discretization of the Up: 4.1 The Finite Element Previous: 4.1.2 Assembly 

4.1.3 Shape Function
The shape function is the function which interpolates the solution between the discrete values
obtained at the mesh nodes. Therefore, appropriate functions have to be used and, as already
mentioned, low order polynomials are typically chosen as shape functions. In this work linear
shape functions are used.

For three­dimensional finite element simulations it is convenient to discretize the simulation
domain using tetrahedrons, as depicted in Figure 4.1. Thus, linear shape functions must be defined
for each tetrahedron of the mesh, in order to apply the Galerkin method described in Section 4.1.1.

Figure 4.1: Finite element mesh of a three­dimensional interconnect structure discretized with tetrahedrons.

Consider a tetrahedron in a cartesian system as depicted in Figure 4.2(a). The linear shape function
of the node   has the form [153]

(4.15)

where  . The coefficients,   and   for each nodal basis function of the


tetrahedral element can be calculated considering the condition [152]

(4.16)

As a result, a system of   equations for the   unknown coefficients is obtained. This procedure


has to be repeated for all tetrahedrons of the mesh, so that the basis functions of all grid nodes are
determined. Furthermore, in order to obtain the discrete system of equations (4.9), the shape
functions have to be derived and integrated, as shown by (4.11) and (4.12).

   
(a)                                                                   (b)
Figure: Tetrahedral finite element. (a) Original coordinate system. (b) Transformed coordinate system.

The calculations can be significantly simplified by carring out a coordinate transformation. A
tetrahedron in a transformed coordinate system is shown in Figure 4.2(b). Each point   of
the tetrahedron in the original coordinate system can be mapped to a corresponding point 
in the transformed coordinate system [155]

     

  (4.17)

     

which in matrix form leads to the Jacobian matrix

(4.18)

In this way, the nodal basis functions for the tetrahedron in the transformed coordinate system are
given by [155]
These shape functions are rather simple, so that the derivatives and integrals required for the finite
element formulation can be readily evaluated in the transformed coordinate system. Given a
function  , the gradient in the transformed coordinates is of the form

(4.20)

where the derivatives are calculated via the chain rule by

These equations can be expressed in matrix notation as

(4.22)

or

(4.23)

where   is the transpose of the Jacobian matrix. Thus, the gradient in the original coordinate
system can be calculated using the transformed coordinate gradient by

(4.24)

where  .

Performing such a coordinate transformation significantly simplifies the practical implementation
of the FEM. The nodal shape functions in the transformed coordinates are fixed and known in
advance, thus, it is not necessary to solve the system of equations formed by (4.15) and (4.16) for
each element of the mesh. Only the Jacobian matrix has to be determined, and the required
calculations for the finite element formulation can be easily evaluated.

       
Next: 4.2 Discretization of the Up: 4.1 The Finite Element Previous: 4.1.2 Assembly

R. L. de Orio: Electromigration Modeling and Simulation

You might also like