Recent Advances, Challenges and Future Trends of Machine Learning and Model Free Predictive Control Applied to Power Electronics
Recent Advances, Challenges and Future Trends of Machine Learning and Mode Free Predictive control Applied to Power Electronics
Prof. Felipe Ruiz Allende Universidad San Sebastián, Chile
Prof. Giovanny Sanchez Instituto Politécnico Nacional, Mexico
Prof. Juan Gerardo Avalos Instituto Politécnico Nacional, Mexico
Wed 04-Mar-2026
Abstract:
Recent advances in embedded computing platforms, including FPGAs, DSPs, and heterogeneous SoC architectures, have enabled the real-time implementation of Machine Learning (ML) algorithms in industrial systems. In power electronics, where converters operate under nonlinear dynamics, parameter uncertainty, and rapidly changing conditions, traditional model-based control strategies often face performance and robustness limitations. In this context, Machine Learning and Model-Free Predictive Control (MFPC) have emerged as powerful alternatives. Reinforcement learning and data-driven predictive approaches allow controllers to adapt using real-time measurements without requiring accurate mathematical models, improving flexibility and robustness in complex operating scenarios. This tutorial provides a concise yet comprehensive overview of recent advances, key challenges, and future trends in Reinforcement learning and MFPC applied to power converters. It covers theoretical foundations, implementation constraints, computational trade-offs, and real-time deployment aspects. Special attention is given to robustness, explainability, and industrial feasibility. The objective is to bridge academic research and practical deployment, equipping participants with a structured understanding of how data-driven and model-free strategies are reshaping next-generation power electronic systems.
Topics:
- Introduction and Motivation (Felipe Ruiz)
- Limitations of classical model-based control
- Overview of Machine Learning paradigms: reinforcement learning
- Introduction to Model Predictive Control (MPC) and Model-Free concepts
- Reinforcement Learning Techniques Applied to Power Electronics (Felipe Ruiz)
- Implementation Challenges and Real-Time Considerations (Felipe Ruiz)
- Real-time constraints
- Embedded implementation aspects
- Computational burden, latency, and memory considerations
- Explainability, generalization, and safety issues
- Model-Free Finite Set Predictive Control (Juan Gerardo Avalos)
- Future Trends and Industrial Perspectives (Giovanny Sanchez)
- Emerging directions: physics-informed ML/MPC
Speaker Bio:
Prof. Felipe Ruiz Allende (Member, IEEE) received the M.Sc. degree at Universitat Politecnica de Catalunya, Spain, 2010. PhD double degrees in electronics engineering between Universidad Tecnica Federico Santa Maria, Valparaiso, Chile, and Warsaw University of Technology, Poland, 2023. Currently, he joined the Energy Transition Center, USS, where he is currently a researcher. His current research interests include power electronic modular multilevel converters, solid-state transformers, and the application of Machine Learning, Artificial Intelligence, and data-driven techniques to power electronics, predictive control, and signal processing, with particular emphasis on model-free and adaptive control strategies for grid-connected power converters.
Prof. Giovanny Sanchez received the M.S. degree at Instituto Politecnico Nacional, Mexico, in 2008, and the Ph.D. degree at Universitat Politecnica de Catalunya, Spain, in 2014. Currently, he is an Associate Professor at the Instituto Politecnico Nacional, Mexico. His main research interests include the development of neuromorphic systems, cryptosystems, adaptive signal processing and image processing.
Prof. Juan Gerardo Avalos was born in Mexico in 1984. He received his M.Sc. in Microelectronics from the National Polytechnic Institute, Mexico, in 2010, and his Ph.D. in Electronics and Communications Engineering from the same institution in 2014. From 2011 to 2012, he was a visiting researcher at the Vienna University of Technology, Austria. He is currently a professor in the Research and Postgraduate Section of ESIME Culhuacan at the National Polytechnic Institute, Mexico
Quantum-Resilient Security for Industrial 6G and Cyber-Physical Systems
Quantum-Resilient Security for Industrial 6G and Cyber-Physical Systems
Prof. Abdullah Aydeger Florida Institute of Technology, USA
Dr. Engin Zeydan Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Spain
Fri 06-Feb-2026
Abstract:
Industrial Cyber-Physical Systems increasingly rely on advanced wireless connectivity, edge computing, and distributed control to support automation, robotics, and critical infrastructure. Emerging 6G networks are expected to play a central role in enabling these systems by providing ultra-reliable and low-latency communication across industrial environments. At the same time, advances in quantum computing pose a long-term threat to classical cryptographic mechanisms that currently protect industrial communication and control channels. Since industrial systems often have long operational lifetimes, addressing quantum security risks early is critical.
This tutorial focuses on quantum resilient security mechanisms for industrial Cyber-Physical Systems, with emphasis on post-quantum cryptography and quantum-based key management. The tutorial introduces the fundamental security challenges posed by quantum adversaries and discusses how quantum-safe mechanisms can be integrated into industrial 6G and next-generation industrial communication architectures. A key focus is placed on practical deployment considerations, including computational overhead, memory usage, latency impact, and system reliability in resource-constrained industrial devices. The tutorial combines conceptual foundations with practical demonstrations and case studies using current user equipment platforms as realistic precursors to future industrial 6G devices. The goal is to provide attendees with actionable insight into designing secure and future-proof Industrial Cyber-Physical Systems in the post-quantum era.
Topics:
- Quantum security threats and their impact on industrial Cyber-Physical Systems
- Fundamentals of post-quantum cryptography for industrial and embedded platforms
- Quantum key distribution concepts and applicability to industrial environments
- Integration of quantum resilient security into industrial 6G and Cyber-physical architectures
- Practical demonstrations and performance evaluation on resource-constrained devices
- Open challenges and research directions for quantum-safe industrial systems
Speaker Bio:
Prof. Abdullah Aydeger is currently an assistant professor at the Electrical Engineering and Computer Science Department at FIT. Prior to joining FIT in August 2022, he was an assistant professor at the School of Computing at Southern Illinois University, Carbondale, since 2020. Dr. Aydeger obtained a Ph.D. Degree in Computer and Electrical Engineering from Florida International University in 2020. His research interests are post-quantum cryptography, network security, and virtualization.
Dr. Engin Zeydan received a PhD degree in February 2011 from the Department of Electrical and Computer Engineering at Stevens Institute of Technology, Hoboken, NJ, USA. Since November 2018, he has been with the Services as Networks (SaS) Research Unit of the CTTC working as a Senior Researcher. He was a part-time instructor at the Electrical and Electronics Engineering department of Ozyegin University Istanbul, Turkey, between January 2015 and June 2018. His research areas include data engineering/science for telecommunication networks and network security.