Automated post-processing and scientific visualization workflow for an existing OpenFOAM bow-thruster CFD simulation using Python and ParaView.
The project focuses on Q-criterion vortex visualization, transient flow analysis, total-pressure calculation, and automated moving-slice visualization, with an emphasis on reproducibility and consistent visualization settings.
CFD post-processing can involve repetitive operations such as loading transient results, applying filters, defining slices, adjusting visualization ranges, configuring cameras, and exporting results.
This project uses Python scripting in ParaView to automate these operations for an existing bow-thruster OpenFOAM dataset.
The workflow was developed to investigate:
- blade-tip vortex structures;
- hub and junction vortex structures;
- transient evolution of the visible Q-criterion field;
- total-pressure distribution; and
- pressure-field variation using an automated moving slice.
The underlying CFD simulation itself is not included in this repository.
- Python
- ParaView
- ParaView Python API
- OpenFOAM CFD data
- NumPy
- Scientific Visualization
- CFD Post-Processing
The project follows a staged post-processing workflow with separate Q-criterion and total-pressure paths.
Existing OpenFOAM CFD Results
|
v
ParaView Reader
|
v
Domain / Geometry Setup
|
+------+------+
| |
v v
Q-Criterion Pressure Path
Path |
| v
v Total-Pressure
q_range_finder.py Calculation
| |
v v
Q-Range Y-Normal Slice
Calibration |
| v
v Moving Slice
q_criterion_final.py |
| v
v Animation
Final Single-
Time-Step Images
|
v
transient_q_pipeline.py
|
v
Transient Combined
Q-Criterion Analysis
The automated post-processing workflow is organized into four Python/ParaView scripts.
Single-time-step Q-criterion display-range calibration workflow.
The calibration stage evaluates candidate visualization ranges for the blade-tip, hub/junction, and combined regions before fixed display settings are selected.
The final selected Q-criterion display ranges used in the project were:
| Region | Q Display Range |
|---|---|
| Blade tip | 50–600 |
| Hub / junction | 50–1200 |
| Combined | 50–1100 |
Source note: The exact original standalone range-finder source was not available in the retained project files. The version included in this repository is a reconstructed portfolio implementation based on the calibration procedure documented in the project report.
Final single-time-step Q-criterion visualization workflow.
The script uses the calibrated display ranges to produce the final visualization for the selected:
- blade-tip region;
- hub/junction region; or
- combined tip-and-hub region.
The processing pipeline includes domain clipping, Q-criterion calculation from the velocity field, an X-normal slice, spatial masking, fixed visualization settings, and screenshot export.
Transient combined blade-tip and hub/junction Q-criterion post-processing workflow.
Main visualization settings include:
- Q display range: 50–1100
- visualization-suppression threshold: Q = 550
- X-normal slice
- Y-Z visualization plane
- spatial masking
- fixed camera configuration
- fixed colour range
Note: Q = 550 is used as a visualization-suppression threshold for visual clarity. It is not treated as a universal physical vortex boundary or as a numerically validated noise threshold.
Automated total-pressure and moving-slice visualization workflow.
The script calculates total pressure from the supplied pressure and velocity fields and creates a Y-normal X-Z slice that moves through the CFD domain while the transient solution advances.
➡️ View the Python source files
Before the final Q-criterion visualizations were created, candidate display ranges were compared.
The objective was to establish consistent visualization settings instead of relying only on automatic colour rescaling.
The final selected ranges were:
- Blade tip: Q = 50–600
- Hub / junction: Q = 50–1200
- Combined: Q = 50–1100
These ranges are visualization settings selected for this workflow and should not be interpreted as universal vortex-identification thresholds.
The blade-tip region was examined using the Q-criterion to visualize rotation-dominated structures around the propeller-tip region.
Selected display range:
Q = 50–600
The central hub and junction region was examined separately to visualize Q-criterion structures around the bow-thruster assembly.
Selected display range:
Q = 50–1200
The combined tip-and-hub configuration was extended across the available CFD time steps.
The transient visualization uses:
Q display range = 50–1100
with a separate visualization-suppression threshold of:
Q = 550
The slice orientation, spatial-selection logic, camera configuration, and colour range are kept consistent throughout the sequence.
This allows representative transient frames to be compared using the same visualization configuration.
The frames represent different CFD time steps from the same automated processing pipeline rather than independently configured visualizations.
Total pressure is calculated from the supplied pressure and velocity fields using:
p_total = p + 0.5 * rho * |U|^2
where:
rho = 1025 kg/m³
The visualization uses a fixed total-pressure display range of:
70–140 kPa
This provides a complementary visualization of the flow field alongside the Q-criterion analysis.
A Y-normal X-Z slice is automatically moved through the CFD domain from:
Y = +0.3 m → Y = -1.5 m
The flow direction is toward -Y.
The slice position and CFD solution time change together during the animation.
Therefore, the sequence contains both spatial and temporal variation. It should be interpreted as an automated visualization and sampling sweep rather than as a material-surface or wake-tracking algorithm.
➡️ View animation results and technical settings
Instead of manually rebuilding the ParaView pipeline for every visualization, the Python scripts define the important processing and visualization parameters programmatically.
The automated workflow includes:
- loading OpenFOAM results;
- selecting relevant mesh regions;
- restricting the working CFD domain;
- calculating Q-criterion from the velocity field;
- defining cross-sectional slices;
- applying spatial masks;
- controlling visualization thresholds;
- fixing colour-transfer ranges;
- maintaining consistent camera settings;
- calculating total pressure;
- animating a moving pressure slice; and
- processing transient CFD states consistently.
This improves repeatability, parameter traceability, and visualization consistency compared with repeatedly configuring the workflow manually.
bow-thruster-cfd-postprocessing-using-python-paraview/
│
├── README.md
├── .gitignore
│
├── src/
│ ├── README.md
│ ├── q_range_finder.py
│ ├── q_criterion_final.py
│ ├── transient_q_pipeline.py
│ └── pressure_moving_slice.py
│
├── images/
│ ├── README.md
│ ├── q_range_calibration.png
│ ├── tip_vortex.png
│ ├── hub_junction_vortex.png
│ ├── transient_q_frames.png
│ ├── total_pressure.png
│ └── moving_pressure_slices.png
│
├── results/
│ └── README.md
│
└── docs/
└── README.md
Animation files, when included, are stored in the
results/directory.
The project demonstrates how scripted post-processing can make CFD visualization workflows more systematic and reproducible.
The Q-criterion workflow provides a method for examining rotation-dominated structures in selected blade-tip and hub/junction regions.
The calibration stage establishes documented visualization ranges before those settings are applied to the final single-time-step results.
The transient workflow then applies a consistent combined visualization configuration across multiple CFD time steps, allowing visible changes in the retained Q-criterion structures to be compared under the same post-processing settings.
The total-pressure workflow provides a second perspective on the flow field by combining the supplied static-pressure field with the local velocity contribution.
The moving slice extends this visualization through different Y positions in the domain while the CFD solution time also advances.
This project focuses on CFD post-processing, automation, and scientific visualization.
The underlying OpenFOAM simulation was supplied separately and is not included in this repository.
The work should not be interpreted as:
- a mesh-independence study;
- solver validation;
- turbulence-model validation;
- direct cavitation prediction;
- acoustic prediction; or
- complete three-dimensional vortex tracking.
Q display limits, spatial masks, and suppression thresholds are visualization choices and require engineering judgement.
A two-dimensional slice cannot represent the complete three-dimensional vortex topology.
The moving pressure slice changes position while the CFD time advances, so spatial development and temporal variation cannot be separated from that animation alone.
A visually clear CFD result should not by itself be interpreted as numerical or physical validation of the underlying simulation.
- CFD post-processing
- transient-flow visualization
- Q-criterion vortex visualization
- total-pressure analysis
- interpretation of CFD field data
- Python scripting
- ParaView Python automation
- parameterized post-processing workflows
- programmable filtering
- transient visualization automation
- moving-slice animation
- consistent colour-range control
- spatial filtering
- cross-sectional flow visualization
- camera standardization
- transient result comparison
- engineering visualization
- reproducible analysis workflows
- parameter traceability
- automation of repetitive engineering tasks
- technical documentation
- result interpretation with documented limitations
Additional project documentation is available in the docs directory.
The original academic report and presentation are not included in this public repository because they contain personal, academic, and administrative information.
This project was completed as part of a university Software Lab and focuses on applying Python automation and ParaView scripting to CFD post-processing.
The repository has been organized as a technical portfolio project highlighting the computational engineering workflow, source code, visualization results, and reproducibility of the post-processing process.