Introduction

What do you want to do

I want to display the VTK file of the 2D temperature simulation result with jupyter.

how?

Read the VTK file with Python, extract the coordinate data and temperature data, and display them with Matplotlib.

environment

OS: Windows 10 home edition

Tool: Anaconda(Python 3.7.4)

VTK file of simulation results

Using the finite element method free software Freefem ++, the thermal diffusion equation was solved and the resulting VTK file was created. The code of the heat diffusion equation used Heat.edp of the example. I will omit how to install and run Freefem ++. The used VTK file is stored in here.

Installing the Python VTK library

Install using conda.

conda install vtk

import relationship

Import the Numpy, Matplotlib, and VTK libraries required for operation.

import numpy as np
import maptlotlib.pyplot as plt
import vtk
from vtk.util import numpy_support
%matplotlib inline

Read VTK file

Since the VTK file was an unstructured grid, read it with Unstructured.

#File reading
filename = "heat_result.vtk"
reader = vtk.vtkUnstructuredGridReader()
reader.SetFileName(filename)
reader.Update()

Converting Temperature Cell Data to Point Data

The coordinate data of this VTK file is Point Data, and the temperature data is Cell Data. I want to have a one-to-one relationship between the coordinate data and the temperature data, so I convert them all to Point Data.

#Convert from cell data to point data
cell2point = vtk.vtkCellDataToPointData()
cell2point.SetInputData(reader.GetOutput())
cell2point.Update()

Extraction of coordinate and temperature data from the converted data

Read the VTK file and use vtk_support to extract the coordinate data and temperature data from the VTK data.

#Numpy coordinate and temperature data
coord = numpy_support.vtk_to_numpy(cell2point.GetOutput().GetPoints().GetData())
x = coord[:,0]
y = coord[:,1]
z = coord[:,2]#do not use. Since it is the result of a two-dimensional thermal diffusion equation, it is all 0.
#GetAbstractArray(0)Label, GetAbstractArray(1)Contains temperature data.
temperature = numpy_support.vtk_to_numpy(cell2point.GetOutput().GetPointData().GetAbstractArray(1))

Color map display in Matplotlib

Now that we're ready, all we have to do is view it in Matplotlib. Since it is unstructured data, the color map is displayed using trikontourf.

##Color map output
plt.tricontourf(x,y,temperature,levels=15,cmap="jet")
plt.colorbar()

All chords

https://github.com/matsxxx/show_vtk_on_jupyter

import numpy as np
import matplotlib.pyplot as plt
import vtk
from vtk.util import numpy_support
%matplotlib inline
#File reading
filename = "heat_result.vtk"
reader = vtk.vtkUnstructuredGridReader()
reader.SetFileName(filename)
reader.Update()

#Convert from cell data to point data
cell2point = vtk.vtkCellDataToPointData()
cell2point.SetInputData(reader.GetOutput())
cell2point.Update()

#Numpy coordinate and temperature data
coord = numpy_support.vtk_to_numpy(cell2point.GetOutput().GetPoints().GetData())
x = coord[:,0]
y = coord[:,1]
z = coord[:,2]#do not use. Since it is the result of a two-dimensional thermal diffusion equation, it is all 0.
#GetAbstractArray(0)Label, GetAbstractArray(1)Contains temperature data.
temperature = numpy_support.vtk_to_numpy(cell2point.GetOutput().GetPointData().GetAbstractArray(1))

#Color map output
plt.tricontourf(x,y,temperature,levels=15,cmap="jet")
plt.colorbar()

reference

For the VTK class, the [class reference] of c ++ (https://vtk.org/doc/nightly/html/index.html) is open to the public. The structure and name of Python VTK is the same as the c ++ reference, so you can refer to it. The following sites were mainly referred to.

https://vtk.org/doc/nightly/html/classvtkFieldData.html
https://vtk.org/doc/nightly/html/classvtkCellDataToPointData.html
https://stackoverflow.com/questions/23138112/vtk-to-matplotlib-using-numpy
https://matplotlib.org/3.1.1/api/_as_gen/matplotlib.pyplot.tricontourf.html

[PYTHON] Read the VTK file and display the color map with jupyter.