0

is there any way to represent a 3D plot with disordered coordinates?

For Example:

I have done a FE-analysis. There is one beam with two different temperatures on each site. The results from the FEA is a temperature profile over the entire component. I did the analysis with ANSYS, so i am able to get the coordinates of each node in the FE-mesh and the temperature of it.

FE result of ANSYS coordinates of the beam

I want to represent these results in a Latex Document. Therefore i imported the coordinates with its temperatures. This is, what i get so far:

enter image description here

Now i want to create a surface between these coordinates and add a colormap to represent the full result like it is shown in ANSYS. This is my code so far:

\documentclass[border=10pt]{standalone}
\usepackage{pgfplots}
\pgfplotsset{compat=1.8}

\begin{document}
    \begin{tikzpicture}
        \begin{axis}[
                xlabel=X-Coordinate,
                ylabel=Y-Coordinate,
                zlabel=Temperature]
            \addplot3+[only marks] table[
                x=X,
                y=Y,
                z=T,
                col sep=semicolon] {Temperatur_Balken.csv}; 
        \end{axis}
    \end{tikzpicture}


\end{document}

This is the data of the .csv-file:

Knotennummer;X;Y;T
1;0;10;19
2;0;0;19
3;50;0;21
4;10;0;20.302
5;20;0;20.602
6;30;0;20.778
7;40;0;20.903
8;50;10;21
9;40;10;20.903
10;30;10;20.778
11;20;10;20.602
12;10;10;20.302
13;0;5;19
14;5;10;19.976
15;5;0;19.976
16;45;0;20.954
17;50;5;21
18;15;0;20.477
19;10;5;20.302
20;25;0;20.699
21;20;5;20.602
22;35;0;20.845
23;30;5;20.778
24;40;5;20.903
25;45;10;20.954
26;35;10;20.845
27;25;10;20.699
28;15;10;20.477

Thank you!

Best regards Theo

TheoH
  • 1
  • Is this the result you're expecting? It looks weird to me but maybe it's exactly what you want: https://i.stack.imgur.com/AFfSI.png – Alenanno Nov 15 '20 at 12:09
  • Thanks for your answer! Unfortunately, this is not the result i want. I tought, that there is a way, that the coordinates that are closest to each other span a surface, like the first picture in my question. I think, that the main problem is, that the coordinates have no order, so they dont span the surface correct. – TheoH Nov 15 '20 at 12:31
  • I thought so too lol I'll take a deeper look at your coordinates later – Alenanno Nov 15 '20 at 12:59
  • You could sort your table, see e.g. https://tex.stackexchange.com/a/102770/. –  Nov 15 '20 at 14:40

0 Answers0