معرفی
Tom Simpson is a Researcher at ETH Zurich specializing in computational mechanics and data-driven engineering methods. Previously completing his PhD at the Chair of Structural Mechanics (2018-2023), he focuses on developing advanced simulation techniques for complex engineering systems. His work bridges theoretical mechanics with machine learning applications.
His educational background includes a Master's degree in Mechanical Engineering from the University of Sheffield (2018), where his thesis explored manifold learning techniques for nonlinear dynamical systems. During his studies, he gained practical experience at the Advanced Manufacturing Research Centre investigating aerospace alloy machining methods.
Simpson's research centers on reduced order modeling, hybrid simulation techniques, and machine learning applications in structural dynamics. His work particularly emphasizes variational autoencoders for nonlinear systems, parametric dependencies in engineering models, and real-time simulation frameworks. Key application areas include structural health monitoring, offshore wind energy systems, and railway vehicle dynamics.
His publication record demonstrates consistent focus on integrating physics-based modeling with data-driven approaches. Recent work shows increasing sophistication in error bound estimation, probabilistic modeling, and online monitoring-conditioned simulations. The research trajectory reveals progression from foundational nonlinear dynamics to cutting-edge hybrid AI-physics simulation frameworks.
Simpson actively contributes to the DyVirt research project and collaborates extensively with Professor Eleni Chatzi's research group. His work appears in leading engineering journals including Mechanical Systems and Signal Processing, Scientific Reports, and Frontiers in Energy Research, demonstrating both theoretical rigor and practical engineering applications.
