EhD-Thruster-Simulator is an open-source simulation tool for modeling Electrohydrodynamic (EHD) thrusters. These devices leverage electric fields, ion flows, and fluid dynamics to generate thrust, with potential applications in propulsion systems such as drones or aerial vehicles. The simulator integrates physics-based modeling, optimization algorithms, and GPU-accelerated computation to predict thrust performance and optimize thruster designs. It also generates 3D STL models for visualization and manufacturing.

• Physics Simulation: Models electrostatic and fluid dynamic interactions within EHD thrusters, incorporating real-world material properties and environmental conditions.
• Optimization: Employs Bayesian optimization to identify efficient thruster designs based on user-defined payload weights.
• Material and Battery Data: Integrates detailed material properties (e.g., copper, aluminum) and battery specifications (e.g., Samsung 21700 series) for realistic simulations.
• STL File Generation: Produces 3D models of thruster components (e.g., emitter, collector, mounts) for visualization or 3D printing.
• GPU Acceleration: Utilizes TensorFlow with DirectML for hardware acceleration on a variety of GPUs, including AMD, Intel, and NVIDIA. For NVIDIA users, an option to use the standard TensorFlow GPU version is available for potentially better performance.

To set up and run the EhD-Thruster-Simulator, ensure you have Python 3.10.11 or later installed. Follow these steps:

1. Clone the Repository:

   git clone https://github.com/versoindustries/EhD-Thruster-Simulator.git
   cd EhD-Thruster-Simulator

2. Create a Virtual Environment (optional but recommended):

   python -m venv env
   source env/bin/activate  # On Windows, use `env\Scripts\activate`

3. Install Dependencies:

   pip install -r requirements.txt

   The requirements.txt file is configured for the DirectML pipeline and includes:

   • tensorflow-cpu==2.10.1
   • tensorflow-directml-plugin
   • numpy==1.26.4
   • scipy
   • scikit-learn
   • numba
   • numpy-stl
   • tqdm
   • tensorboard
   • matplotlib

This project is built for the TensorFlow DirectML pipeline by default, enabling hardware acceleration on a wide range of GPUs, including AMD, Intel, and NVIDIA. The requirements.txt file includes tensorflow-cpu==2.10.1 and tensorflow-directml-plugin to support this configuration.

TensorFlow DirectML is a plugin that allows TensorFlow to utilize DirectML, a hardware-accelerated machine learning API for Windows. It broadens TensorFlow's compatibility, enabling GPU acceleration on systems without NVIDIA hardware.

If you have an NVIDIA GPU and prefer to use the standard TensorFlow GPU version for potentially better performance, modify the requirements.txt file as follows:

• Remove the line containing tensorflow-directml-plugin.
• Replace tensorflow-cpu==2.10.1 with tensorflow==2.10.1.

After making these changes, reinstall the dependencies using:

pip install -r requirements.txt

Ensure you have the necessary NVIDIA drivers and CUDA Toolkit installed. Refer to the TensorFlow GPU guide for detailed setup instructions.

Note on GPU Usage: This project uses TensorFlow with DirectML by default, supporting hardware acceleration on various GPUs, including AMD, Intel, and NVIDIA. No additional setup is required beyond installing the dependencies. For NVIDIA users who prefer the standard TensorFlow GPU version, follow the instructions in the "TensorFlow Configuration" section above.

The primary script is ehd_simulator.py. Run it from the command line to optimize thruster designs for a specified payload weight.

python ehd_simulator.py

When prompted, enter the payload weight in kilograms (e.g., 1.0). The simulator will perform optimization and save results in the optimization_results directory.

View all options with:

python ehd_simulator.py --help

Customize the simulation by editing:

• materials.json: Material properties (e.g., density, conductivity).
• samsung_21700_batteries.json: Battery specifications (e.g., capacity, voltage).

1. Optimize for a 0.5 kg Payload:

   python ehd_simulator.py
   # Enter "0.5" when prompted

   Results, including STL files, are saved in optimization_results/payload_0.500kg.

2. View Generated STL Files: Find STL files (e.g., collector_tube.stl, emitter.stl) in the design subdirectories. Use software like Blender or MeshLab to visualize them.

3. Modify Material Properties: Edit materials.json to adjust properties, then rerun the simulation:

   "copper": {
       "density_kg_m3": 8960,
       "elec_conductivity_S_m": 5.96e7,
       ...
   }

• Memory Usage: High-resolution grids can be memory-intensive. Ensure sufficient RAM and GPU VRAM (e.g., 8GB+ recommended).
• Computation Time: Optimization duration varies with grid size and iterations. Reduce grid resolution for faster initial tests.
• Caching: Intermediate results are cached in the simulation_cache directory. Ensure adequate disk space.

Detailed information on simulation models, optimization algorithms, and code structure is available in the documentation.

specs code was built on, could be ran cpu on less with much longer runtimes.

• Ryzen 7 2700x
• 64 GB RAM
• AMD rx6600

Contributions are welcome! To contribute:

1. Fork the repository.
2. Create a feature branch (git checkout -b feature-name).
3. Commit your changes (git commit -m "Add feature").
4. Push to the branch (git push origin feature-name).
5. Open a pull request.

Please follow the Code of Conduct and report issues via GitHub Issues.

There are currently several variations of our end cap we are prototyping with right now, will update the code with the most efficient model. Top ring for the wire is not that great of a mechanical design, as learned from the first prototype. Update coming with a much cleaner design.

This project is licensed under the Apache License 2.0. See the LICENSE file for details.

• Benefits:
  • Recognition on the project's website and documentation.
  • Access to project updates and newsletters.
• Description:
  • Ideal for individuals or small organizations who want to support the project without needing additional perks.
• Sponsor Now

• Benefits:
  • All Supporter Tier benefits.
  • Early access to new features and releases (e.g., beta versions or private repository access).
  • Standard support for issues and feature requests.
• Description:
  • Suitable for users or companies who actively use the simulator and want to stay ahead of updates.
• Sponsor Now

• Benefits:
  • All Contributor Tier benefits.
  • Priority support for issues and bug fixes.
  • Feature request prioritization (requests considered in the next development cycle).
  • Logo displayed on the project's homepage.
• Description:
  • Designed for companies or organizations that rely heavily on the simulator and want their needs addressed promptly.
• Sponsor Now

• Benefits:
  • All Patron Tier benefits.
  • Guaranteed inclusion of one feature request per year (subject to feasibility and project maintainers' approval).
  • Participation in quarterly strategy meetings with project maintainers to discuss roadmap and strategic decisions.
  • Invitation to an Advisory Board to provide input on major project directions.
• Description:
  • For organizations that want a significant say in the project's future while ensuring their use cases are prioritized.
• Sponsor Now

• Benefits:
  • All Director Tier benefits.
  • Opportunity to sponsor a major project milestone or release (e.g., funding a specific version or feature set).
  • Custom development work on a specific module or feature (within the scope of the project and open-source principles).
  • Co-branding opportunities (e.g., company name/logo prominently featured in project materials).
• Description:
  • For major stakeholders who want to deeply integrate the project into their operations and ensure it aligns with their long-term goals.
• (Note: Sponsorship at this level is by arrangement. Please contact us directly.)

• Email: [email protected]
• Website: www.versoindustries.com (currently not live)
• Commercial Licensing: See COMMERCIAL-LICENSE.md

https://discord.gg/pBrSPbaMnM

If you use this simulator in your research, please cite it as:

Verso Industries, Michael B. Zimmerman. (2025). EhD-Thruster-Simulator. GitHub repository, https://github.com/versoindustries/EhD-Thruster-Simulator

• Thanks to the developers of TensorFlow, NumPy, SciPy, scikit-learn, trimesh, and shapely for their foundational libraries.
• Special appreciation to Verso Industries and Michael B. Zimmerman for initial development efforts.