Sustainability Research in High-Speed Trains: A Neural Network-Driven Approach

Strategic Insights -

Researchers at Polytechnic University Milan have proposed a novel approach to reducing energy consumption and carbody vibration in high-speed trains through the use of a convolutional neural network (CNN) driven control strategy. The study was funded by the National Key Research and Development Program of China and the H2020 Marie Sklodowska-Curie Actions. The researchers developed a co-simulation platform combining multibody dynamics simulation software and MATLAB/Simulink to analyze the effects of train velocity, track curvature, and scale factor of the Skyhook controller on energy efficiency and lateral carbody vibration. A CNN was constructed to predict energy consumption and riding comfort under complex operation scenarios, and a bi-objective optimization model was developed to adjust the scale factor of the Skyhook controller according to different running conditions.

Key Takeaways:

  • Researchers at Polytechnic University Milan proposed a CNN-driven control strategy to reduce energy consumption and carbody vibration in high-speed trains.
  • The strategy combines multibody dynamics simulation software and MATLAB/Simulink to analyze the effects of train velocity, track curvature, and scale factor on energy efficiency and lateral carbody vibration.
  • The CNN was trained on simulation data to predict energy consumption and riding comfort under complex operation scenarios.
  • The bi-objective optimization model was developed to adjust the scale factor of the Skyhook controller according to different running conditions.
  • The optimization results indicate that energy consumption can be reduced by up to 15.90% and lateral vibration by up to 47.78% through the employment of the proposed control strategy.
  • The study was funded by the National Key Research and Development Program of China and the H2020 Marie Sklodowska-Curie Actions.
  • The research has implications for the sustainable development of high-speed rail transportation systems.

Statistics:

  • 15.90% reduction in energy consumption through the proposed control strategy.
  • 47.78% reduction in lateral vibration through the proposed control strategy.
  • 101183 DOI for the journal article "A convolutional neural network driven suspension control strategy to enhance sustainability of high-speed trains."
  • The study involved a team of researchers from Polytechnic University Milan, including Duo Zhang, Hong-Wei Li, Fang-Ru Zhou, Yin-Ying Tang, and Qi-Yuan Peng.

Sources:

  • A convolutional neural network driven suspension control strategy to enhance sustainability of high-speed trains, published in Energy Conversion and Management: X (2025), 27():101183.
  • National Key Research and Development Program of China.
  • H2020 Marie Sklodowska-Curie Actions.
  • Polytechnic University Milan.