Organizers: Prof. Dr. Mohammad Mohammadi Aghdam, Mechanical Engineering Department, Amirkabir University of Technology, Tehran , Iran
Dr. Ali Fallah, Assistant Professor, Department of Automotive Engineering, Atılım University, Ankara, Türkiye
Physics-informed machine learning (PIML) has emerged as a powerful paradigm for solving and analyzing partial differential equations governing complex engineering systems by integrating physical laws directly into learning architectures. Among these methods, Physics-Informed Neural Networks (PINNs) and related operator-learning frameworks provide mesh-free, data-efficient alternatives to classical numerical techniques, while preserving physical consistency and interpretability.
This symposium aims to bring together researchers working at the intersection of numerical analysis, applied mathematics, and computational mechanics to present recent advances in physics-informed machine learning methods. Particular emphasis will be placed on the numerical analysis aspects of these approaches, including formulation, stability, convergence behavior, error assessment, and benchmarking against established numerical methods such as finite element, finite difference, and spectral techniques.
Contributions addressing applications in solid and structural mechanics, heat and mass transfer, fluid mechanics, wave propagation, additive manufacturing, and multiphysics or nonlinear systems are especially welcome. The symposium seeks to foster dialogue between the numerical analysis and engineering communities, promoting the development of reliable, physics-consistent, and scalable machine-learning-based solvers for real-world mechanical and multiphysics problems.