A5047626644- Magnetic confinement fusion research
- Fusion materials and technologies
- Nuclear reactor physics and engineering
- Laser-Plasma Interactions and Diagnostics
- Superconducting Materials and Applications
- Nuclear Physics and Applications
- Ionosphere and magnetosphere dynamics
- Time Series Analysis and Forecasting
- Chaos control and synchronization
- Particle accelerators and beam dynamics
- Evolutionary Algorithms and Applications
- Complex Systems and Time Series Analysis
- Atomic and Subatomic Physics Research
- Atmospheric and Environmental Gas Dynamics
- Nonlinear Dynamics and Pattern Formation
- Anomaly Detection Techniques and Applications
- Advanced Optical Sensing Technologies
- Evolution and Genetic Dynamics
- Neural Networks and Applications
- Fault Detection and Control Systems
- Computational Physics and Python Applications
- Medical Imaging Techniques and Applications
- High-Energy Particle Collisions Research
- Nuclear Engineering Thermal-Hydraulics
- Metaheuristic Optimization Algorithms Research
University of Rome Tor Vergata
2015-2024
National Agency for New Technologies, Energy and Sustainable Economic Development
2012-2024
Culham Science Centre
2014-2020
Royal Military Academy
2020
National Institute for Laser Plasma and Radiation Physics
2019
Max Planck Society
2018
Max Planck Institute for Plasma Physics
2018
Abstract The JET hybrid scenario has been developed from low plasma current carbon wall discharges to the record-breaking Deuterium-Tritium plasmas obtained in 2021 with ITER-like Be/W wall. development started pure Deuterium refinement of current, and toroidal magnetic field choices succeeded solving heat load challenges arising 37 MW injected power ITER like environment, keeping radiation edge core controlled, avoiding MHD instabilities reaching high neutron rates. have re-run Tritium...
Abstract ITER is of key importance in the European fusion roadmap as it aims to prove scientific and technological feasibility a future energy source. The EUROfusion consortium labs within Europe contributing preparation exploitation operation aspires exploit outcomes view DEMO. paper provides an overview major progress obtained recently, carried out frame new (initiated 2021) work-package called ‘ Pr eparation I TER O peration’ (PrIO). directly supported by eleven PrIO contributions given...
Detecting disruptions with sufficient anticipation time is essential to undertake any form of remedial strategy, mitigation or avoidance. Traditional predictors based on machine learning techniques can be very performing, if properly optimised, but do not provide a natural estimate the quality their outputs and they typically age quickly. In this paper new set tools, probabilistic extensions support vector machines (SVM), are introduced applied for first JET data. The output constitutes...
Abstract This paper reports the first experiment carried out in deuterium–tritium addressing integration of a radiative divertor for heat-load control with good confinement. Neon seeding was time D–T plasma as part second campaign JET its Be/W wall environment. The technical difficulties linked to re-ionisation heat load are reported T and D–T. compares impact neon on plasmas their D counterpart detachment, localisation radiation, scrape-off profiles, pedestal structure, edge localised modes global
Notwithstanding the efforts exerted over many years, disruptions remain a major impediment on route to magnetic confinement reactor of tokamak type. Machine learning predictors, relying adaptive strategies, have recently proved achieve very good performance. Even if their last generation implement 'from scratch' approach learning, i.e. they can start predicting after first example each class (safe and disruptive), it would be extremely useful profit from experience previous devices, when new...
Abstract In many scientific applications, it is important to investigate how certain properties scale with the parameters of systems. The experimental studies scalings have traditionally been addressed log regression, which limits results power laws and theoretical not data-driven dimensionless quantities. This has also case in nuclear fusion, scaling energy confinement time a crucial aspect understanding physics transport design future devices. Traditionally two main assumptions are at...
For many years, machine learning tools have proved to be very powerful disruption predictors in tokamaks.On the other hand, vast majority of techniques deployed assume that input data is independent and sampled from exactly same probability distribution for training set, test set final real time deployment.This hypothesis certainly not verified practice, since experimental programmes evolve quite rapidly, resulting typically ageing consequent suboptimal performance.This paper describes...
Abstract In the JET DTE2 campaign a new method was successfully tested to detect heating of bulk electrons by α-particles, using dynamic response electron temperature T e modulation ion cyclotron resonance (ICRH). A fundamental deuterium (D) ICRH scheme applied tritium-rich hybrid plasma with D-neutral beam injection (NBI). The i and accelerated deuterons leads modulated α -heating large delay respect other terms. significant phase ∼40° is measured between central , which can only be...
The total emission of radiation is a crucial quantity to calculate the power balances and understand physics any Tokamak. Bolometric systems are main tool measure this important physical through quite sophisticated tomographic inversion methods. On Joint European Torus, coverage bolometric diagnostic, due availability basically only two projection angles, limited, rendering very ill-posed mathematical problem. A new approach, based on maximum likelihood, has therefore been developed...
The most widely used forms of model selection criteria, the Bayesian Information Criterion (BIC) and Akaike (AIC), are expressed in terms synthetic indicators residual distribution: variance mean-squared error residuals respectively. In many applications science, noise affecting data can be expected to have a Gaussian distribution. Therefore, at same level error, models, whose more uniformly distributed, should favoured. degree uniformity quantified by Shannon entropy. Including entropy BIC...
Many measurements are required to control thermonuclear plasmas and fully exploit them scientifically. In the last years JET has shown potential generate about 50 GB of data per shot. These amounts require more sophisticated analysis methodologies perform correct inference various techniques have been recently developed in this respect. The present paper covers a new methodology extract mathematical models directly from without any priori assumption their expression. approach, based on...
The extrapolation of the energy confinement time to next generation devices has been investigated both theoretically and experimentally for several decades in tokamak community. Various scaling expressions have proposed using dimensional dimensionless quantities. They are all based on assumption that scalings power law form. In this paper, an innovative methodology is extract tokamaks directly from experimental databases, without any previous about mathematical form scalings. approach obtain...
To produce fusion reactions efficiently, thermonuclear plasmas have to reach extremely high temperatures, which is incompatible with their coming into contact material surfaces. Confinement of using magnetic fields has progressed significantly in the last years, particularly tokamak configuration. Unfortunately, all devices, and metallic ones, are plagued by catastrophic events called disruptions. Many disruptions preceded anomalies radiation patterns, ITER-relevant scenarios. These specific...
The next generation of Tokamak devices is expected to work at very high radiated fractions, well above 90%, preserve the integrity plasma facing components in general and divertor particular. In addition maintaining confinement, these configurations will also have guarantee a low disruptivity. An accurate determination emitted radiation therefore become increasingly important, not only for global power balances but specific regions cross section (for example properly control detachment)....
Abstract Determination of causal-effect relationships can be a difficult task even in the analysis time series. This is particularly true case complex, nonlinear systems affected by significant levels noise. Causality modelled as flow information between systems, allowing to better predict behaviour phenomenon on basis knowledge one causing it. Therefore, theoretic tools, such transfer entropy, have been used various disciplines quantify causal relationship events. In this paper, Transfer...
In recent years, a new tomographic inversion method based on the Maximum Likelihood (ML) approach has been adapted to JET bolometry. Apart from its accuracy and reliability, key advantage is ability provide reliable estimates of uncertainties in reconstructions. The original algorithm was implemented validated using MATLAB software tool. This work presents accelerated version compatible ITER fast controller platform with Ubuntu 18.04 or Codac Core System distributions (6.1.2). C++...
Understanding the details of correlation between time series is an essential step on route to assessing causal relation systems. Traditional statistical indicators, such as Pearson coefficient and mutual information, have some significant limitations. More recently, transfer entropy has been proposed a powerful tool understand flow information signals. In this paper, comparative advantages entropy, for determining horizon influence, are illustrated with help synthetic data. The technique...
This work describes the behaviour of global energy and particle confinement on JET observed in a massive database H-mode plasmas covering almost whole lifetime operations, both with carbon metal wall. The analysis is focused type I ELMy H-modes stationary phases. It shown that plasma density regime determined mainly by current, edge safety factor, triangularity last closed flux surface hydrogenic isotope mass. That consistent for regardless divertor configuration or facing materials. On...
The inadequacies of basic physics models for disruption prediction have induced the community to increasingly rely on data mining tools. In last decade, it has been shown how machine learning predictors can achieve a much better performance than those obtained with manually identified thresholds or empirical descriptions plasma stability limits. main criticisms these techniques focus therefore two different but interrelated issues: poor “physics fidelity” and limited interpretability....
Access to the H mode of confinement in tokamaks is characterized by an abrupt transition, which has been subject continuous investigation for decades. Various theoretical models have developed and multi-machine databases experimental data collected. In this paper, a new methodology reviewed scaling laws temperature threshold access mode. The approach based on symbolic regression via genetic programming allows first extraction most statistically reliable from available data. Nonlinear fitting...
Abstract Control of instabilities such as ELMs and sawteeth is considered an important ingredient in the development reactor-relevant scenarios. Various forms ELM pacing have been tried past to influence their behavior using external perturbations. One main problems with these synchronization experiments resides fact that are periodic or quasi-periodic nature. Therefore, after any pulsed perturbation, if one waits long enough, always bound occur. To evaluate effectiveness techniques, it...