Lecturer: M. A. Mendez, Associate Professor at the von Karman Institute for Fluid Dynamics (VKI).
Registration: Prior registration required.
Registration here.
Description: Real-time and adaptive modeling lies at the heart of digital twinning, one of the most transformative concepts in engineering and applied sciences. Digital twins seek to virtually replicate physical systems using models that continuously “learn” from data, automating data collection, validation, and refinement to achieve a self-learning status. This enables real-time prediction, decision-making, and control. The framework relies on mathematical tools that optimally blend imperfect physical models with noisy and partial observations.
This lecture provides an introduction to two fundamental sets of methods: statistical approaches for state estimation and variational techniques for parameter identification. After a brief recap of essential concepts from probability, calculus, and optimization, we will cover the core methods in each category, namely the Extended Kalman Filter (EKF) and adjoint-based optimization. These topics will be explored through a series of hands-on exercises in Python. Participants are expected to have a working Python environment installed in advance to fully benefit from the practical sessions.
Course Plan (June 2nd, 2025, 10.00 – 14.00, room to be confirmed)
- Lecture 1 : Introduction and Fundamentals (1 h)
- Lecture 2 : Linear and Extended Kalman Filter (1h 30)
- Lecture 3: The Adjoint Method for Parameter Identification (1h30)
BIO M. A. Mendez is an Associate Professor at the von Karman Institute for Fluid Dynamics (VKI), where he teaches courses on modelling and control of fluid flows, measurement techniques, signal processing and machine learning within the Research Master program. He currently also visiting professor at the Universite’Libre de Bruxelles and at the Universidad Carlos III de Madrid. At VKI, he is the head of the Machine Learning 4 Fluid Systems (ML4F) group at the von Karman Institute, working on data-driven methods for experimental fluid mechanics, flow control and modelling of fluid flows, and organizer of several VKI Lecture series on these topics such as “Hands on Machine Learning for Fluid Dynamics”, now approaching its sixth edition. He is the recipient of an ERC Starting Grant for the Re-Twist Project, aiming to develop digital twin technologies for a wind turbines, drones and cryogenic propellant storage systems and the main editor of the book ‘Data – Driven Fluid Mechanics’, published by Cambridge University Press.
PhD in Aerospace Engineering UC3M 