Erase cell from mesh by index

Question

Dear all,

I would like to know if there a method/command to erase a cell by index.

I would like to remove a group of cell that failed in my stress criteria for example and run this model again without these elements.

Is anyone has any idea?

Best Regards

Answer 1

I don't exactly understand what you try to achieve but one way to remove a cell is to create a new mesh based on the old one without the cell you want to remove. Here is a 1D example:

from dolfin import * 
import numpy as np

mesh = UnitIntervalMesh(5)

old_cells = mesh.cells()
old_coors = mesh.coordinates()

new_coors = np.delete(old_coors, (3), axis=0)
new_cells = np.delete(old_cells, (-1), axis = 0)

new_mesh = Mesh() 
editor = MeshEditor() 
editor.open(new_mesh, 1, 1)
editor.init_vertices(len(new_coors))
editor.init_cells(len(new_cells))

vert_id = 0
for vert in new_coors:
    editor.add_vertex(vert_id, vert)
    vert_id += 1

cell_id = 0
for c in range(len(new_cells)): 
    editor.add_cell(cell_id, new_cells[c][0], new_cells[c][1])
    cell_id += 1 

editor.close()

plot(new_mesh, interactive=True)

Comments

Thank you for your answer, but I am looking for easier command, for example:

I have four cells:

cell_ids = [ 0, 1, 2, 3, 4 ]

and I would like to delete the cell ids 1 and 2 from mesh object.

I am looking for a command like this for example:

mesh.delete_cells([1,2])

and then I will run the model again.