cv

VISIT PhD Student

University of Washington

VISIT PhD student in the Department of Electrical and Computer Engineering. My research focuses on SHallow REcurrent Decoders for state estimation in nuclear reactors, under the supervision of prof. J. Nathan Kutz.

• Machine Learning
• Reduced Order Modelling
• Python
• Coupled PDEs

Teaching Assistant

Politecnico di Milano

Teaching assistant for the course of Fission Reactor Physics I, taught by prof. Antonio Cammi. My duties include holding theoretical and exercise sessions.

• Neutron Transport
• Neutron Dynamics
• Reactor Kinetics
• Python

Teaching Assistant

Politecnico di Milano

Teaching assistant for the course of Fission Reactor Physics I, taught by prof. Antonio Cammi. My duties include holding exercise sessions in preparation of the written test.

• Neutron Transport
• Neutron Dynamics
• Reactor Kinetics
• Python

PhD Student

Politecnico di Milano

PhD student in Energy and Nuclear Science and Engineering. My research interests include reduced order modelling and machine learning for nuclear reactors.

• Nuclear Reactors
• Machine Learning
• Reduced Order Modelling
• Python
• Data Assimilation

Adolescents’ Educator

Oratorio San Luigi

Adolescents’ Educator

Oratorio San Luigi

Actor for no-profit theatre

Depramirao

Football Coach

Stella Azzurra '56, Oratorio San Luigi

Master of Science

Politecnico di Milano, Italy

Nuclear Engineering

• Fission Reactor Physics
• Atomic and Nuclear physics
• Nuclear Instrumentation
• Reliability, safety and risk analysis
• Design and modelling of nuclear reactors
• Computational Fluid Dynamics

Bachelor of Science

Politecnico di Milano, Italy

Energy Engineering

• Calculus
• Physics (mechanics, thermodynamics, electromagnetism)
• Solid mechanics
• Fluid Mechanics
• Applied Thermodynamics
• Energy Conversion and Fluid machines
• Analytical and numerical methods for engineering

Best Master Thesis in Engineering

Cultural Association CISE2007

Best Paper award

ICAPP-2023 Conference

Best Student Paper award

NUTHOS-14 Conference

Multi-physics model bias correction with data-driven reduced order techniques: Application to nuclear case studies

Applied Mathematical Modelling

This paper investigates the use of data-driven reduced-order modeling (ROM) techniques for model bias correction in multi-physics nuclear reactor simulations. It focuses on correcting inaccuracies in models by incorporating real-time data to improve the efficiency and accuracy of nuclear reactor simulations.

Application of a non-intrusive reduced order modeling approach to magnetohydrodynamics

Physics of Fluids

The study applies reduced-order modeling (ROM) techniques to magnetohydrodynamics (MHD) simulations, demonstrating how ROM can improve computational efficiency while preserving accuracy. The focus is on simulating MHD flows in the context of fusion reactors and validating the ROM approach through case studies.

Impact of Malfunctioning Sensors on Data-Driven Reduced Order Modelling: Application to Molten Salt Reactors

EPJ Web Conf

This work explores the impact of malfunctioning sensors on data-driven reduced-order modeling (ROM) for Molten Salt Reactors. It examines how sensor failures affect the performance of ROM techniques and proposes strategies to mitigate their effects in reactor monitoring and control.

Real-Time State Estimation of Neutron Flux in Molten Salt Fast Reactors from Out-core Sparse Measurements

arXiv

This preprint discusses a method for real-time state estimation of neutron flux in Molten Salt Fast Reactors using sparse out-core measurements. The study introduces a shallow recurrent decoder to improve state estimation in reactors with limited sensor coverage.

OFELIA: An OpenMC-FEniCSx coupling for neutronic calculation with temperature feedback

Nuclear Engineering and Design

The article presents OFELIA, a multi-physics tool coupling OpenMC Monte Carlo simulations for neutron physics with FEniCSx for thermal-hydraulic analysis. The coupling strategy is applied to determine the temperature profile of a PWR fuel pin, and the results are compared with literature data to validate the approach.

Data-driven model order reduction for sensor positioning and indirect reconstruction with noisy data: Application to a Circulating Fuel Reactor

Nuclear Engineering and Design

This work applies data-driven model order reduction (MOR) to optimize sensor positioning and reconstruct non-observable fields in the core of a Circulating Fuel Reactor (CFR). The study uses reduced-order modeling techniques to handle noisy data and demonstrates their applicability in reactor safety and monitoring.

A finite element implementation of the incompressible Schrödinger flow method

Physics of Fluids

This article explores a finite element implementation of the incompressible Schrödinger flow method, a novel approach for simulating inviscid fluids. The paper presents case studies and shows that this method may offer computational advantages over traditional methods in fluid dynamics, especially for complex geometries and real-world applications.

Hybrid Data Assimilation methods, Part II: Application to the DYNASTY experimental facility

Annals of Nuclear Energy

This paper investigates the application of Hybrid Data Assimilation (HDA) methods to the DYNASTY experimental facility, aiming to combine Model Order Reduction techniques with Data Assimilation to estimate system states from sparse measurements in nuclear reactor systems. The methods are tested on experimental data and validated for real-world applications in the context of natural circulation.

Hybrid data assimilation methods, Part I: Numerical comparison between GEIM and PBDW

Annals of Nuclear Energy

This paper investigates hybrid data assimilation methods by comparing the Generalized Empirical Interpolation Method (GEIM) and the Parameterized-Background Data-Weak (PBDW) approach on a 3D Backward Facing Step airflow benchmark. The study highlights their effectiveness in noisy data environments, with GEIM outperforming PBDW in terms of lower reconstruction error and fewer sensors required.

Non-intrusive system state reconstruction from indirect measurements: A novel approach based on Hybrid Data Assimilation methods

Annals of Nuclear Energy

This work introduces a two-step hybrid data assimilation approach for real-time system state estimation using reduced order modeling. It extends traditional frameworks to handle cases where variables of interest are not fully measurable, showcasing its effectiveness on a thermal hydraulics case and the Molten Salt Fast Reactor.

Stabilization of Generalized Empirical Interpolation Method (GEIM) in presence of noise: A novel approach based on Tikhonov regularization

Computer Methods in Applied Mechanics and Engineering

The Empirical Interpolation Method (EIM) and its generalized version (GEIM) optimize sensor placement and real-time state estimation in thermo-hydraulic systems. Tikhonov regularization is used to stabilize these methods and improve reconstruction accuracy when data is noisy.