Course Description:
Machine Learning (ML) and Artificial Intelligence (AI) are transforming fluid dynamics, enabling faster simulations, smarter turbulence modeling, and innovative flow-control strategies. For CFD engineers and students, upskilling in AI is essential to stay competitive, accelerate research, and contribute to the next generation of engineering solutions.
In this course, you will learn the fundamentals and the new trends in machine learning for fluid dynamics and engineering. With a practical style, combining master classes and code development, you will receive training and support from two world-leading scientists in the field. The course covers basic neural networks, physics-informed architectures, spatial and temporal predictions, control, reduced-order models, and much more.
Key information:
- Course Instructor: Prof Ricardo Vinuesa and Prof Sergio Hoyas
- Course content: 10 Lessons (~15 hours of content), including theory and practical sessions involving coding and implementation.
- Active discussion forum: You can directly interact with Prof. Ricardo Vinuesa and Prof. Sergio through the discussion forum and clear all your doubts and questions
- Total access period: 12 Months from the day of enrollment
- Access to the course: Once you make the payment, your login ID and password will be sent automatically via email.
- How to access: Once enrolled, you can immediately access by logging in to the portal
Key feature:
Active discussion with experts: Directly interact with the instructors, Prof. Ricardo and Prof. Sergio, via the discussion forum to clarify all your questions
Instructors
Prof. Ricardo Vinuesa
Professor Ricardo Vinuesa is a leading researcher at the intersection of fluid dynamics, artificial intelligence (AI), and sustainability. He is currently an Associate Professor of Aerospace Engineering at the University of Michigan, USA.
He studied Mechanical Engineering at the Polytechnic University of Valencia (UPV) in Spain, followed by an MSc and PhD in Mechanical and Aerospace Engineering at the Illinois Institute of Technology (IIT), Chicago. Before joining Michigan, he was an Associate Professor at KTH Royal Institute of Technology in Stockholm, where he also served as Vice Director of the KTH Digitalization Platform and Lead Faculty at the KTH Climate Action Centre.
Professor Vinuesa’s research combines high-fidelity simulations with machine learning methods, including deep neural networks, reinforcement learning, and explainable AI, to advance turbulence prediction, optimization, and flow control. His influential Nature Communications (2020) paper on AI and the UN Sustainable Development Goals demonstrated both the opportunities and risks of AI for global sustainability.
He has published over 200 papers and received prestigious honors, including a European Research Council (ERC) Consolidator Grant, the Göran Gustafsson Award, and the IIT Outstanding Young Alumnus Award.
Prof. Sergio Hoyas Calvo
Professor Sergio Hoyas Calvo is a Professor of Aerospace Engineering at the Universitat Politècnica de València (UPV) in Spain. He holds a degree and PhD in Mathematical Sciences from the Complutense University of Madrid, completing his doctorate in 2003. His academic career includes research and teaching roles at several Spanish universities, and he also serves as a Lecturer in Mathematics internationally.
Prof. Hoyas is internationally recognized for his expertise in turbulence, fluid mechanics, computational fluid dynamics (CFD) and the application of high-performance computing and numerical methods to complex mathematical and engineering problems. He has published widely, including highly cited work on turbulent flows and wall-bounded turbulence, and he has contributed research integrating modern tools such as explainable AI to advance understanding of turbulent dynamics.
Course content in detail
- Theory Topics:
-
Lesson 1: Introduction to Machine Learning and Multilayer Perceptrons (MLP)
(Foundations of ML and Feedforward Neural Networks) -
Lesson 2: Recurrent Neural Networks (RNNs) and Long Short-Term Memory Networks (LSTM)
(Time-Series Modeling and Temporal Dynamics in Fluid Systems) -
Lesson 3: Physics-Informed Neural Networks (PINNs)
(Embedding Governing Equations into Neural Networks) -
Lesson 4: Convolutional Neural Networks (CNNs)
(Spatial Feature Extraction for Flow Fields and Images) -
Lesson 5: Generative Adversarial Networks (GANs) and Explainable Artificial Intelligence (XAI)
(Data Generation and Model Interpretability in Engineering Applications) -
Lesson 6: Deep Reinforcement Learning (DRL)
(Control, Optimization, and Flow Control Strategies) -
Lesson 7: Reduced-Order Models (ROMs), Autoencoders (AEs), and Transformers
(Model Compression, Latent Representations, and Attention Mechanisms for Fluid Systems)
-
- Practical Topics (Including coding):
-
Practical Lesson 1: Implementing Multilayer Perceptrons (MLP) and Recurrent Neural Networks (RNNs) for Engineering Data
-
Practical Lesson 2: Data-Driven Modeling of the Lorenz System (Nonlinear Dynamical Systems Case Study)
-
Practical Lesson 3: Advanced Implementation of MLPs and RNNs for Time-Dependent Fluid Flow Prediction
-
- Certificate
- Earn a Certificate of Completion by successfully finishing the course and submitting all assignments and code projects.
Frequently asked question
- Do I get a certificate?
Yes, once you complete the lessons and assignments, a course completion certificate with the instructor’s name will be issued. - What if I don’t understand some portion or need to clarify some doubts?
You can ask all your questions directly to the instructors via the discussion forum - Should I know programming to learn this course?
Basic knowledge of Python will be useful - Is there any prerequisite?
Knowledge of CFD, Fluid dynamics, and coding with Python will be useful
Who is this course for?
- Practicing CFD Engineers who want to upskill and stay relevant in the job market
- Bachelor students (Mechanical, Aerospace, Chemical, Civil Engineering)
- Researchers, master’s students, and PhD students
- AI and ML enthusiasts who want to upskill
Other details
- Total access to the course: 1 year. Further extension can be provided upon request.
- Computer requirements: Minimum 4 GB RAM and i3 processor
Course Includes
- 12 Lessons
Ratings and Reviews
Dr.-Ing. in computational engineering, implementing SFEM and LSFEM from scratch in Python. I enrolled to bridge classical FEM with modern AI/ML — specifically PINNs. My goal: translate weighted LSFEM residual strategies into weighted PINN loss functions. This course is exactly the bridge I was looking for.