Enhancing nonlinear dynamics analysis of railway vehicles with artificial intelligence: a state-of-the-art review

Railway vehicle dynamics involves modelling, simulating, and analysing the motion and interaction of rail vehicles under external force, exhibiting numerous nonlinear behaviours. Artificial intelligence (AI) is increasingly crucial in analysing and understanding these behaviours. This review presents a comprehensive survey of the latest research on AI-enhanced nonlinear dynamics simulation and analysis of railway vehicles, focusing on a few essential topics such as data-driven modelling and simulation, dynamics performance prediction, and control. The benefits and limitations of using AI to enhance the nonlinear dynamics analysis of railway vehicles are discussed in detail. The results highlight the significant potential of incorporating AI into classical nonlinear dynamics simulation and analysis methods. This provides critical insights for researchers and practitioners seeking to advance the state-of-the-art in railway vehicle dynamics. Furthermore, the review addresses the limitations and challenges associated with using AI in this field, including concerns about the quality of training data, the complexity of neural networks, and the interpretability of black-box models. These considerations ensure the reliability and applicability of AI-enhanced models and methods. Looking ahead, the review identifies and suggests future research directions, such as data accumulation and annotation, hybrid dynamics models, applications of transfer learning and deep reinforcement learning algorithms. These directions aim to further enhance the accuracy, efficiency, and reliability of AI-enhanced models and methods in railway vehicle dynamics.