If I have understood your question correctly:

1. Do topology optimization.

2. Select only those elements with densities you are interested in

3. Extract the surface details of those elements

4. Using the nodes, you generate tessellated surfaces (bunch of triangles)

5. Export the tessellated surfaces as IGES or STEP or STL into your CAD package of choice

6. Using these imported surface as an outline, reconstruct a proper structure following the broad contours of the tessellations.

I have done it many times in the past successfully. All you would need is a simple Python or C or anything code to process the node and element connectivity data, create the tessellations and export to CAD.

Nicolò Lodini You might be interested in PERMAS. All kinds of optimization are available. The so-called Design Wizard in VisPER will create a smooth hull. Export to .stl is supported as well. A free education edition is available.

Try below steps

1. Assuming you have loaded necessary data like 'densities', 'ConnectivityMatrix', 'NodeCoordinates'

2. Define the density threshold for your surface (e.g., 0.3)

densityThreshold = 0.3;

3. Identify elements with densities meeting the threshold

elementsToPlot = find(densities >= densityThreshold);

4. Extract element connectivity and node coordinates for selected elements

selectedConnectivity = ConnectivityMatrix(elementsToPlot, :);

selectedNodes = NodeCoordinates(unique(selectedConnectivity(:)), :);

5. Create a surface mesh using the selected elements and nodes (Use MATLAB's patch or triangulation functions to create the surface)

6. Assuming 'Faces' contains connectivity information and 'Vertices' contains node coordinates

patch('Faces', selectedConnectivity, 'Vertices', selectedNodes, 'FaceColor', 'g');

7. Export the surface as an STL file (Replace 'surface.stl' with desired file name) stlwrite('surface.stl', Faces, Vertices);

% Example assuming you have 'Coordinates', 'ConnectivityMatrix', and 'Density' loaded

isovalue = 0.3;

% Identify elements with density above the isovalue

selectedElements = find(Density > isovalue);

% Extract coordinates and connectivity for selected elements

selectedCoordinates = Coordinates';

selectedConnectivity = ConnectivityMatrix(selectedElements, :);

% Create isosurface

figure;

h = isosurface(selectedCoordinates(:, 1), selectedCoordinates(:, 2), selectedCoordinates(:, 3), Density, isovalue);

% Display the surface

p = patch(h);

isonormals(selectedCoordinates(:, 1), selectedCoordinates(:, 2), selectedCoordinates(:, 3), Density, p);

set(p, 'FaceColor', 'red', 'EdgeColor', 'none');

daspect([1 1 1]); axis tight;

view(3); camlight; lighting gouraud;

% Save as STL

stlwrite('output_file.stl', p.Faces, p.Vertices);

Try this

I get this error: V must be a 3D array

Thank you