Airfoil Self-Noise Data Set
Start something that matters
Predict the self-generated noise of an airfoil blade encountering smooth flow.

Learn more and start today

Airfoil Self-Noise Data Set

The NASA data set comprises different size NACA 0012 airfoils at various wind tunnel speeds and angles of attack. The span of the airfoil and the observer position were the same in all of the experiments.
  • Attribute Information:
  • 1. Frequency, in Hertz.
  • 2. The angle of attack, in degrees.
  • 3. Chord length, in meters.
  • 4. Free-stream velocity, in meters per second.
  • 5. Suction side displacement thickness, in meters.
  • 6. Scaled sound pressure level, in decibels.

Challenge

This dataset is linear or no?

What is the best dimensionality reduction? PCA or T-SNE?

Are there empty or null values?

Model

This dataset was trained with linear models, non-linear models and a neural network.

Linear models: Linear regression, Lasso, Ridge, ElasticNet, BayesianRidge

Non-linear models: SVR

Neural network: Hidden layers with Tensorflow and Keras

To the code

Result

See the results for each model:

Model R2 score MSE score
Neural network 88.92% 5.40
SVR 79.81% 9.85
BayesianRidge 45.68% 26.52
LinearRegression 45.68% 26.53
Ridge 45.68% 26.52
ElasticNet 45.55% 26.59
Lasso 45.55% 26.59