diff --git a/examples/013_thermal_example.py b/examples/013_thermal_example.py index 53b98b683..0cbdf0e7f 100644 --- a/examples/013_thermal_example.py +++ b/examples/013_thermal_example.py @@ -26,15 +26,17 @@ Thermal analysis ---------------- -PyDPF Composites can also be used to post-process thermal analyses. -In this case, the simulation is a two-step analysis where the results of -a thermal analysis are an input of the structural analysis. So, the RST -contains temperature and structural results. -The example mimics a PCB which was modeled with Ansys Composites PrePost (ACP). -where the solid model feature of ACP is used to generate the volume mesh. +This example shows how to evaluate a thermal analysis. +The simulation's thermal analysis results are also +the input of a structural analysis. +Therefore, the RST file contains temperature and structural results. -In detail, the example shows how to extract the temperatures for a specific ply, -and a specific material. +The example imitates a printed circuit board (PCB) that was +modeled with Ansys Composites PrepPost (ACP), using the solid model feature to +generate the volume mesh. + +Descriptions of how to extract temperatures for a specific ply +and material are provided. .. note:: @@ -67,22 +69,22 @@ # %% # Initialize the model # ~~~~~~~~~~~~~~~~~~~~ -# The composite model is initialized with the composite files and the server. -# It provides access to the mesh, results, lay-up and materials +# Initialize the composite model with the composite files and server. +# The model provides access to the mesh, results, lay-up, and materials. composite_model = CompositeModel(composite_files, server) # %% -# Get Results - Temperatures -# ~~~~~~~~~~~~~~~~~~~~~~~~~~ -# The temperatures are stored under structural_temperature +# Get temperature results +# ~~~~~~~~~~~~~~~~~~~~~~~ +# The temperatures are stored under ``structural_temperature``. temp_op = composite_model.core_model.results.structural_temperature() temperatures_fc = temp_op.outputs.fields_container() # %% -# Ply-wise results -# ~~~~~~~~~~~~~~~~ -# Ply-wise results can be easily extracted using the function -# :func:`.get_ply_wise_data` and by passing the ply name. +# Get ply results +# ~~~~~~~~~~~~~~~ +# Extract the ply-wise results by passing the ply name +# to the function :func:`.get_ply_wise_data`. all_ply_names = get_all_analysis_ply_names(composite_model.get_mesh()) print(all_ply_names) @@ -99,10 +101,10 @@ composite_model.get_mesh().plot(nodal_values) # %% -# Material-wise results -# ~~~~~~~~~~~~~~~~~~~~~ -# It is also possible to filter the results by material. -# In this example the element-wise maximum temperature +# Get material results +# ~~~~~~~~~~~~~~~~~~~~ +# You can filter the results by material. +# In this example, the element-wise maximum temperature # is extracted for the material `Honeycomb Aluminum Alloy`. print(composite_model.material_names) material_id = composite_model.material_names["Honeycomb Aluminum Alloy"]