Export Nodal results time history via Python API

[a1rodrig]

Dear STKO,

I am currently trying to export nodal displacement time history to a text file.
I am able to do so by going over the displacement at each time step following the Python API tutorials provided.

My question is:
Is it possible to access the entire displacement time-history for a single node at once?
I have over 9000 time-steps and 30 nodes, therefore going step by step in a 'for loop' becomes a very lengthy process

I have attached my code for reference.
Thank you for your help.

[STKO Team]

Try this:

Extract_Nodal_Results.zip

(1.93 KiB) Downloaded 253 times

I fixed an error in the indexing. The step_id is not necessarily 1-based (it depends on how many stages you have).
I used the zip command to iterate over the step_id and the step_time at the same time:

Code: Select all

for step_id, step_time in zip(all_steps, all_times): #Loop for timesep

To speed up the process I made a reduced mesh containing only the nodes you need to extract.
It will reduce the size of the dataset of each evaluation:

Code: Select all

# make a fake mesh with only the required nodes
# 1. do a first evaluation to get the real mesh
# 2. from the real mesh get the requested nodes and copy them into the reduced mesh
# 3. save the reduced mesh in the evaluation options
opt.step = all_steps[0]
field = Disp.evaluate(opt)
reduced_mesh = MpcMesh()
for i in NODES:
	inode = field.mesh.getNode(i)
	reduced_mesh.addNode(inode)
opt.mesh = reduced_mesh

Let me know the speed-up you get. I tried it on a small model and it already gave a 20x speed-up

[a1rodrig]

Reducing the mesh for each evaluation worked perfectly!
It was able to speed up the process by 20x

Thank you very much for your help.

Andres R

[STKO Team]

You're welcome. And the speed up will be greater as the ratio total_number_of_nodes/required_number_of_nodes increases (i.e. for large models the speed up is larger)