ML Based Design


Currently aircraft wing design is based on human calculations and experience. It is a labour intensive process and can take a very long time to achieve the best results.


The design process is more automated and new wing design iterations can be produced in a much shorter time frame, hence achieving the optimal design selection quicker than ever before.

The How

  1. Use CAE software to generate a large amount of design data
  2. Develop a B-spline model to extract features from the geometry data
  3. Develop a Support Vector Machine (SVM) regression model
  4. Applying Bayesian optimisation to the SVM model leads to the best design feature