Bonassi Fabio
Post-doctoral researcher at Uppsala University
Systems and Control Division
Dep. of Information Technology
Uppsala University
Sweden
I’m a post-doctoral reasearcher at the Uppsala University, Sweden. My research interests lie at the intersection of Deep Learning, dynamical system modeling and optimization-based control.
With my research activities, I try to bridge the gaps between the deep learning community and the System & Controls community, showing that the use of Recurrent Neural Networks (RNNs) for system indentification and MPC design can combine performances and safety.
I am also interested in the application of deep learning and optimization-based control approaches to the Power System.
Reconciling Deep Learning and Control Theory
As a PhD student at Politecnico di Milano, Italy, I have been investigating the use of Recurrent Neural Networks for data-driven control of unknown dynamical systems. We investigated:
- The Input-to-State Stability (ISS) and Incrementally Input-to-State Stabile (𝛿ISS) of RNNs.
- Training strategy for learning RNNs with stability and robustness certifications.
- RNN-based Model Predictive Control (MPC) and Internal Model Control (IMC) schemes.
This ideas have been described in my PhD dissertation, defended on February 2023.
Optimization-based control for the smart grid
I am also interested in the application of optimization-based control algorithms in the context of the smart grid. During my research activities, I have worked on the following topics:
- Design of MPC architectures for the optimal energy management and coordination of microgrids & prosumers.
- Analysis of the European regulatory framework for the participation of prosumer aggregators in the electricity markets.
- Design of optimization-based energy management architectures for aggregators of microgrids & prosumers.
news
Mar 27, 2024 | The article Structured state-space models are deep Wiener models has been accepted for the 20th System Identification Symposium (SYSID 2024)! |
---|---|
Nov 01, 2023 | The paper Nonlinear MPC design for incrementally ISS systems with application to GRU networks has been published on Automatica, and it is available on publisher’s website. |
Aug 15, 2023 | I am honored to announce that I received the Dimitris N. Chorafas PhD Award. |
Jul 03, 2023 | The paper Robust offset-free nonlinear model predictive control for systems learned by neural nonlinear autoregressive exogenous models has been published on the International Journal of Robust and Nonlinear Control. The article is available in open access at the publisher website. |
Jun 19, 2023 | On June 19, 2023, I started my position as a Postdoctoral Researcher in Machine Learning for Control at Uppsala University, Sweden, in the Systems and Control Division. During this activity I will collaborate with Prof. Thomas Schön and Prof. Per Mattsson. |
selected publications
- PhDReconciling deep learning and control theory: recurrent neural networks for model-based control designFeb 2023