Introduction: Welcome to our Airfoil Prediction project! We have developed different regressor model to predict the coefficient of lift-to-drag ratio (CI/CD) of airfoils based on their design parameters. After testing these models, we found that the Support Vector Regression (SVR) and MULTI algorithm outperforms the others and is the best choice for future use. The project aims to predict the CI/CD ratio using machine learning models. The dataset contains features such as Alpha, Reynolds, and Airfoil name and output as CI/CD ratio. Six machine learning models are used to predict the CI/CD ratio. The models are Linear Regression, RANDOM FOREST, SVR, Multi Perceptron, and Support Vector Regression (SVR).

Observation: After comparing the performance of all models, it is observed that the SVR model and Multi provides the best accuracy and is the most suitable model for future use.

Jupyter Notebook: The Jupyter Notebook of last part is final part for you above code parts is optional.

In the third part, a function is created to predict the CI/CD ratio for a given set of input parameters using the best-tuned SVR model.

Conclusion: The project successfully predicted the Airfoil Name and CI/CD ratio using machine learning models. The best-performing model is the Multi model. The code can be modified and used for future applications. Scatter Plot To visualize the performance of the models, a scatter plot was created comparing the actual and predicted values of the ci/cd ratio. The scatter plot for the random forest regressor model is shown below:

For any further inquiries or questions, please feel free to contact us via email at [email protected] o SAAD ALI

Best regards,

MR SAAD