Roberto Rocchetta

Roberto Rocchetta is a researcher within the Inteligent energy system group (SUPSI-DACD-ISAAC). He holds a Master in Energy engineering from Bologna (IT), a PhD in Reliability engineerign Liverpool (UK) and he worked in prestigious institutes and universities such as NIA/NASA-Langley (Virginia USA) and TU/Eindhoven. Roberto's research focuses on decision-making under risk and uncertainty, inteligent energy systems menagement and optimization, ucertainty quantification and reliability/risk based design. Rocchetta’ s research is highly multidisciplinary and combines ideas from system reliability engineering, statistical learning theory, stochastic optimization, uncertainty quantification and machine learning. He is the first author of more than 10 peer-reviewed journal articles and more than 15 conference papers. His works, despite being relatively new (2018-2021), already gathered more than 380 citations and had a striking impact on the research community.

I hold a Master and a Bachelor degree in Energy Engineering from the University of Bologna (Italy), a Master of Research in Decision-making Under Risk and Uncertainty from the University of Liverpool. I received my Ph.D. from the University of Liverpool, United Kingdom 2019, while doing research at the Institute for Risk and Uncertainty and School of Engineering. My dissertation dealt with the development of computational frameworks for the analysis of resilience, reliability and vulnerability analysis of power grid systems ( Download Dissertation Here ). During my Ph.D. and Masters I was a visiting student at ETH Zurich, Switzerland, and at Polytechnic of Milano, Italy, and at Ecole Central de Paris, France, where he spent two five and six months, respectivelly. I worked in the start-up group ARAMIS s.r.l.in Milan Italy, where I helped developping Machine Learning (Deep Reinforcement Learning) framework to tackle health menagement and mainteinance problems.

My research focuses on developing advanced uncertainty quantification framework for the analysis of complex systems and critical infrastructures. I have experience with optimization under uncertainty and reliability-based design-optimization (RBDO). More recently, I developed anomaly detection methods based on SVM ensembles equipped with formal generalization error bounds. I am also interested in several machine learning methods applicable to anomaly detection, remaining useful life estimation and tasks related to the topic "Prognostics and Health Management". During my PhD, my research focused on power networks resilience and reliability. Specifically, I developed computational tools for the analysis/identification of risks and criticalities based on topological analysis of the network structure and physics-based simulation of the network reaction to critical failures (e.g. cascading events, targeted attacks).