This was developed primarily for the purposes of visualizing 2D unsteady fluid dynamics and heat transfer simulations.
We use Python-VTK to convert VTK data to NumPy data, and visualizes 2D data with Matplotlib.
Naturally convecting water in a cavity is initially at steady state, with a hot boundary on the left, cold boundary on the right, and adiabatic top and bottom walls. The cold wall temperature is dropped well below the freezing temperature, initiating freezing from right to left.
The black streamline thickness is proportional to the velocity. This is convenient for visualizing flow within the liquid part of phase-changing domains, where the velocity in the solid region is nearly zero. A white contour line was added to mark the freezing temperature contour, separating the liquid and solid regions.
Our code writes the PNG images at each time step using the same approach as in the test__plot_unsteady_superposed_scalar_field_and_streamlines test,
though this is a higher resolution version of the water freezing test data.
We used a separate converter to generate a GIF.