Deadline Date: 31 May 2027
Modern engineering systems—spanning civil infrastructure, aerospace structures, offshore platforms, and energy facilities—operate under increasingly demanding and uncertain dynamic environments. Accurately characterizing their nonlinear responses, coupled multi-physics behavior, and failure processes remains a fundamental challenge. Classical analytical methods and conventional finite element approaches, while well-established, struggle with high-dimensional parameter spaces, real-time constraints, and the intricate interplay of physical mechanisms that govern complex engineering dynamics.
The convergence of advanced numerical methods and physics-informed machine learning (PIML) offers a transformative opportunity to address these challenges. Techniques such as physics-informed neural networks (PINNs), operator learning, and multi-fidelity data fusion embed governing physical laws directly into learning architectures, enabling reliable predictions with sparse or noisy data. Coupled with high-performance computing and digital twin platforms, these methods are redefining how engineers simulate, monitor, and optimize structural systems under complex dynamic loading conditions.
This Special Issue on "Advanced Numerical Methods and Physics-Informed Machine Learning for Complex Engineering Dynamics" seeks original contributions that advance the frontiers of computational mechanics and intelligent structural analysis.
Topics of interest include: (1) PIML and PINNs for structural dynamics and wave propagation; (2) multi-scale and multi-physics numerical methods; (3) data-driven and physics-hybrid surrogate modeling; (4) digital twins for real-time structural health monitoring; (5) probabilistic methods and uncertainty quantification; (6) intelligent optimization and reliability-based design; (7) fluid–structure interaction and coupled dynamic systems; and (8) numerical methods for advanced and composite materials. Original research articles, comprehensive reviews, and engineering case studies are all welcome.
Editorial board
