Tuesday, January 20, 2026 | 5:00 PM – 6:00 PM
EN:lab, Department of Energy – Politecnico di Milano, Campus Bovisa
Sommario
Join us for the fifth seminar in our Meet the STEP-CHANGErs series, a platform showcasing the innovative research of PhD students in the Science, Technology, and Policy for Sustainable Change (STEP) program. Each session explores cutting-edge sustainability challenges and solutions, offering insights into how emerging research shapes real-world systemic change. The seminars also provide a valuable networking opportunity, with an aperitif following each session.
This session features two exciting talks:
Guido Carlo Masotti – Development of a Techno-Economic Methodology for Flexible Nuclear Hybrid Energy Systems
As the share of variable renewable energy grows and hard-to-decarbonize sectors demand low-carbon solutions, flexible energy systems are increasingly vital. Nuclear Hybrid Energy Systems (NHES) — combining nuclear reactors with other energy sources, storage, and non-electric applications — offer significant potential but face complex technical, economic, and regulatory challenges. This research develops a comprehensive framework to assess Small Modular Reactor-driven NHES, from system design and operation to their role in long-term decarbonization. By combining physics-based models, dynamic simulations, and techno-economic optimization, the study evaluates cost-effective layouts, operational strategies, and policy-relevant insights to support future NHES deployment.
Filippo Dainelli – XAI-GPI: An Interpretable and Adaptive Machine Learning Genesis Index for Tropical Cyclones
Predicting tropical cyclone formation remains a major scientific challenge, particularly across different ocean basins. This research introduces XAI-GPI, a data-driven, explainable machine learning framework that estimates annual cyclone formation while clarifying the environmental factors driving it. Applied across six tropical basins, XAI-GPI outperforms traditional Genesis Potential Indices in capturing interannual variability, while remaining physically interpretable. The work highlights the influence of key factors such as vertical wind shear, humidity, and large-scale climate variability, demonstrating how explainable AI can enhance understanding of tropical cyclones and inform climate resilience strategies.