Scribd
Upload
21 views

Jupyter Lab-Course

The document describes the steps to implement a linear isotropic electrostatic finite element method (FEM) and validate the results. It involves: 1) Building a global matrix with Dirichlet boundary conditions and source vector with known potentials; 2) Inverting the matrix and solving the system to obtain unknown potentials; 3) Comparing results to a past paper for validation. It then describes calculating potentials and electric fields on elements to visualize the field, including functions to determine potentials and electric fields at points.

Uploaded by

Juan David Carrillo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
21 views

Jupyter Lab-Course

The document describes the steps to implement a linear isotropic electrostatic finite element method (FEM) and validate the results. It involves: 1) Building a global matrix with Dirichlet boundary conditions and source vector with known potentials; 2) Inverting the matrix and solving the system to obtain unknown potentials; 3) Comparing results to a past paper for validation. It then describes calculating potentials and electric fields on elements to visualize the field, including functions to determine potentials and electric fields at points.

Uploaded by

Juan David Carrillo
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Linear Isotropic Electrostatic FEM implementation – Part III

SYSTEM RESOLUTION & RESULTS VALIDATION


1. Define the non-singular global matrix Creso:
(a) Build a NxN identity matrix (N → number of nodes);
(b) Analyse input data to know which node possesses a Dirichlet BC (fixed potential value);
(c) Mount the resolution global matrix accordingly.
2. Recover the “right-side” source array V_known:
(a) Analyse input data and recover known potentials;
(b) Mount the source vector accordingly.
NOTE: both processes 1) and 2) can be realized simultaneously
3. Verify that Creso is invertible and then calculate the inverse of the matrix. Analyse obtained results.
4. Resolve the system [Creso] V_unkown = V_known.
5. Compare obtained results with Sadiku’s 1989 paper (validation).

POTENTIAL AND ELECTRIC FIELD CALCULATION & VISUALIZATION


1. Recover local potential values for each element.
2. Create a function potential(x,y) that:
(a) Take as input the coordinates (x,y) of the point you want to calculate the potential for;
(b) Calculate the values of interpolation functions ϕi(x,y) {i=1,2,3} for M elements of our domain;
(c) Analyse the values of ϕi(x,y) {i=1,2,3} in each of those and decide wether or not point (x,y)
belong to element 1, 2, … or M;
(d) Calculate the potential V(x,y) and return it.
3. Define input variables to be used with the previous function.
4. Create a function elec_field() that:
(a) Creates a Mx2 matrix to store Ex and Ey for each element;
(b) Calculate the electric field value in each element.
5. Plot and analyse the vector field obtained with the elec_field() function.

TRY YOUR CODE WITH DIFFERENT DATA AND VALIDATE IT!

You might also like