Scribd
Upload
4 views

report on sbfem coding

The document discusses the author's understanding of MATLAB coding related to the Scaled Boundary Finite Element Method (SBFEM), focusing on Laplace interpolants and one-dimensional shape functions. The author expresses confusion regarding the derivation of shape functions for 4-node elements, the generalization to more nodes, and specific variables and modifications in the stiffness matrix computation. Additionally, questions are raised about the handling of Jacobian determinants and the dimensions of matrices in relation to the number of nodes and shape function degrees.

Uploaded by

bibeksnm
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
4 views

report on sbfem coding

The document discusses the author's understanding of MATLAB coding related to the Scaled Boundary Finite Element Method (SBFEM), focusing on Laplace interpolants and one-dimensional shape functions. The author expresses confusion regarding the derivation of shape functions for 4-node elements, the generalization to more nodes, and specific variables and modifications in the stiffness matrix computation. Additionally, questions are raised about the handling of Jacobian determinants and the dimensions of matrices in relation to the number of nodes and shape function degrees.

Uploaded by

bibeksnm
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/ 2

REPORT ON MY UNDERSTANDING OF THE MATLAB CODING ON SBFEM

Laplace Inetrpolant:

From the matlab code of Laplace Interpolant, I focused on understanding the concept of loop over

vertices and circum function. The code iterates through each vertex of the polygon. The shape

function values are normalized by the sum of numerators in order to ensure the partition of unity.

Laplace interpolant is fundamentally based on the concept of distances to the circumcircles of

triangles formed by neighbouring nodes in the polygon. In order to construct the shape function,

the distance to the circumcircles are calculated.

One Dimensional Shape Functions:

This matlab code interprets the function that computes shape function and its derivatives for a

finite element of 2, 3, 4 nodes in 1D space. In my understanding, each nodes represents an

increasing order like 2 nodes for linear shape function, 3 nodes for quadratic shape function, 4

nodes for cubic or higher orders.

My confusion lies on the shape function for 4 node element. How are shape function

expressions derived, for the 4-node element?

Also, If I want to generalize this to 5 or more nodes, are there any specific principles or

methodologies that I should follow?

Scaled Boundary Finite Element Method:

I understand the general mathematical approach the code implements for the stiffness matrix

computation. The code structure and purpose are clear to me. However, I have some confusions

about specific variables, the role of eigenvector modification.


 The variables like subecon and xyc are used to define sub element connectivity and

coordinates but are not defined in the function inputs. My confusion is should these be pre-

computed?

 I understand that the eigenvalue problem is used to compute the scaled boundary

parameters. However, the modification of the eigenvector matrix (v(:,end-1:end) = 0) is

unclear to me. Why and how this step is being used?

 How do the dimensions of these matrices change with the number of nodes or the degree

of the shape functions? Should we be considering any constraints to ensure these matrices

are well-formed?

 How does the code handle cases where the Jacobian determinant in the given code becomes

small or negative?

These are some of the part where I am confused on, Professor. Currently, I am also working on

understanding the phase field paper. Thank you.

You might also like