Researchers Develop Novel Approach for Predicting Supersonic Aircraft Aerodynamics Using Machine Learning

Strategic Insights -

A team of researchers at Zhejiang University has published a new study proposing a novel approach that integrates machine learning algorithms with computational fluid dynamics (CFD) simulations to efficiently predict the aerodynamic performance of supersonic aircraft under cruising flight conditions. This innovative method has been shown to be more efficient than traditional CFD simulations, providing faster results without sacrificing accuracy. The research was supported by the National Key Research and Development Program of China and the Fundamental Research Funds for the Central Universities.

Key Takeaways:

  • The researchers developed a novel approach that integrates machine learning algorithms with CFD simulations to predict supersonic aircraft aerodynamics.
  • The proposed method employs Bayesian optimization algorithms to enhance prediction accuracy and has been shown to be more efficient than traditional CFD simulations.
  • The study evaluated three intelligent models (XGBoost, Bagging, and Random Forest) based on their principles, advantages, and limitations and applied them to predict the aerodynamic performance of a reference supersonic passenger aircraft.
  • The research concluded that machine learning-based prediction methods can provide faster results without sacrificing accuracy and offers deeper insights into the aerodynamic optimization of supersonic aircrafts.
  • Shapley Additive explanation was applied to the optimal model, improving interpretability and offering deeper insights into the aerodynamic optimization of supersonic aircrafts.
  • The study was supported by the National Key Research and Development Program of China and the Fundamental Research Funds for the Central Universities.
  • Qijia Liu, Jiechao Zhang, Yao Zheng, and Yaolong Liu were the authors of the study published in AIP Advances.

Statistics:

  • The proposed method is more efficient than traditional CFD simulations, providing faster results without sacrificing accuracy.
  • The XGBoost model has the highest predictive accuracy for lift and drag coefficients.
  • The Bagging model is excellent at predicting the pitch moment coefficient.

*Machine learning-based prediction methods offer faster results without sacrificing accuracy, improving the predictability of supersonic aircraft aerodynamics.

  • The study was published in AIP Advances, Volume 15, Issue 8, 2025, Article Number 085111.

Sources:

  • "Supersonic aircraft aerodynamic performance prediction based on machine learning approach." AIP Advances, 2025, 15(8), 085111-085111-11.
  • Zhejiang University, Hangzhou, People's Republic of China.
  • National Key Research and Development Program of China.
  • Fundamental Research Funds for the Central Universities.