Ana Larrañaga

Biography

Ana Larrañaga is an Acting Assistant Professor in the Department of Mechanical Engineering at the University of Washington, where she is also affiliated with the NSF AI Institute in Dynamic Systems.

Previously, she held a postdoctoral research position at the University of Washington working with Professor Steven L. Brunton on topics related to active learning and digital twins. She received her Ph.D. in Energy Efficiency and Sustainability in Engineering and Architecture from the University of Vigo and the University of the Basque Country in 2024, graduating summa cum laude. Her doctoral work focused on developing data-driven surrogate models for complex optimization problems in enhanced heat transfer applications. She also holds an M.Sc. in Industrial Mathematics, specializing in Numerical Simulation, and a B.Sc. in Energy Engineering, specializing in Energy Efficiency, both completed summa cum laude. Her academic background includes computational fluid dynamics, combustion, scientific machine learning, and numerical modeling of engineering systems. 

Education

• Ph.D. in Energy Efficiency and Sustainability in Engineering and Architecture, University of Vigo & Basque Country, 2024.
• M.Sc. in Industrial Mathematics, specialized in Numerical Simulation, University of Santiago de Compostela, A Coruña, Vigo, Carlos III & Polytechnic of Madrid, 2021.
• B.Sc. in Energy Engineering, specialized in Energy Efficiency, University of Vigo, 2019.

Previous appointments

• Postdoctoral Scholar in Mechanical Engineering (UW) and NSF AI Institute in Dynamic Systems, 2025.

Research Statement

Dr. Larrañaga research focuses on developing interpretable and generalizable data-driven models for complex multidisciplinary, multiscale, and multimodal engineering systems. She combines scientific machine learning, reduced-order modeling, sparse sensing, active learning, and constrained optimization to build surrogate models that integrate physical knowledge with simulation and experimental data of varying fidelity. A central goal of her work is to understand when and why these models fail to generalize, and how physics-informed learning and adaptive sampling can improve predictive accuracy while reducing the computational cost of engineering design. Her research is motivated by applications in sustainable aviation and aerodynamics, particularly aerodynamic shape optimization, sustainable fuel formulation, combustion, fluid mechanics, and heat transfer. 

Select publications

1. A. Larrañaga, S. L. Brunton, J. Porteiro, and D. Lopez-Pintor. The Fuel Optimizer: A Data-Driven Numerical Framework for Formulation of Aviation Turbine Fuel. Under revision in Applied Energy, 2026.
2. A. Larrañaga, U. Fasel, and S. L. Brunton. How Low Can You Go? Active Learning for Sparse Model Discovery in the Ultra-Low-Data Limit. Under revision in Nature Communications, 2026.
3. Y. C. Li, A. Larrañaga, S. L. Brunton, and U. Fasel. An Introduction to Sparse Identification of Nonlinear Dynamics for Engineering Applications. Under review in SIAM Review, Education section, 2026.
4. A. Larrañaga, J. Martínez, J. L. Míguez, and J. Porteiro. Robust Optimization of Heat-Transfer-Enhancing Microtextured Surfaces Based on Machine Learning Surrogate Models. International Communications in Heat and Mass Transfer, 151, 107218, 2024.
5. A. Larrañaga, S. L. Brunton, J. Martínez, S. Chapela, and J. Porteiro. Data-Driven Prediction of the Performance of Enhanced Surfaces from an Extensive CFD-Generated Parametric Search Space. Machine Learning: Science and Technology, 4(2), 025012, 2023.