Welcome to the Airfoil Flow Analysis project repository! This software computes essential aerodynamic coefficients for airfoil profiles using a robust panel method approach in Python. It accurately predicts Cp (Coefficient of pressure), Cl (lift), and Cd (drag), and provides insightful visualizations of velocity streamlines.
- Panel Method Implementation: Utilizes a subdivision technique for precise aerodynamic calculations.
- Geometry Generation: Includes an embedded NACA airfoil generator for versatile shape configurations.
- Integration with XFoil: Seamless integration with XFoil for enhanced accuracy and validation.
- Streamline Visualization: Generates graphical representations of flow streamlines for intuitive analysis.
To complement the software's accuracy, a small-scale wind tunnel has been constructed. This setup includes:
- Test Section Specifications: 0.3m x 0.3m with a 9:1 contraction ratio.
- Measurement Capabilities: Arduino-based system for real-time data logging of lift, drag, and Cp polars.
- Experimental Validation: Validates software predictions against empirical data with a high degree of correlation (within 15% error margin).
For more details on the wind tunnel design and integration, visit ldak.dev/projects/wind-tunnel.
Continued efforts are focused on:
- Enhancing Computational Efficiency: Optimizing code for faster calculations and expanded modeling capabilities.
- Improving Accuracy: Refining algorithms for more precise aerodynamic predictions across various airfoil shapes.
Feel free to explore and contribute!