Search results for "Machine"
showing 10 items of 2592 documents
Deep learning for core-collapse supernova detection
2021
The detection of gravitational waves from core-collapse supernova (CCSN) explosions is a challenging task, yet to be achieved, in which it is key the connection between multiple messengers, including neutrinos and electromagnetic signals. In this work, we present a method for detecting these kind of signals based on machine learning techniques. We tested its robustness by injecting signals in the real noise data taken by the Advanced LIGO-Virgo network during the second observing run, O2. We trained a newly developed Mini-Inception Resnet neural network using time-frequency images corresponding to injections of simulated phenomenological signals, which mimic the waveforms obtained in 3D num…
Single trajectory characterization via machine learning
2020
[EN] In order to study transport in complex environments, it is extremely important to determine the physical mechanism underlying diffusion and precisely characterize its nature and parameters. Often, this task is strongly impacted by data consisting of trajectories with short length (either due to brief recordings or previous trajectory segmentation) and limited localization precision. In this paper, we propose a machine learning method based on a random forest architecture, which is able to associate single trajectories to the underlying diffusion mechanism with high accuracy. In addition, the algorithm is able to determine the anomalous exponent with a small error, thus inherently provi…
3D-Printable Model of a Particle Trap: Development and Use in the Physics Classroom
2019
Quadrupole ion traps are modern and versatile research tools used in mass spectrometers, in atomic frequency and time standards, in trapped ion quantum computing research, and for trapping anti-hydrogen ions at CERN. Despite their educational potential, quadrupole ion traps are seldom introduced into the physics classroom not least because commercial quadrupole ion traps appropriate for classroom use are expensive and difficult to set up. We present an open hardware 3D-printable quadrupole ion trap suitable for the classroom, which is capable of trapping lycopodium spores. We also provide student worksheets developed in an iterative design process, which can guide students while discovering…
An energy residual-based approach to gradient effects within the mechanics of generalized continua
2012
AbstractGeneralized continua exhibiting gradient effects are addressed through a method grounded on the energy residual (ER)-based gradient theory by the first author and coworkers. A main tool of this theory is the Clausius-Duhem inequality cast in a form differing from the classical one only by a nonstandard extra term, the (nonlocality) ER, required to satisfy the insulation condition (its global value has to vanish or to take a known value). The ER carries in the nonlocality features of the mechanical problem through a strain-like rate field, being the specific nonlocality source, and a concomitant higher-order long-range stress (or microstress) field. The thermodynamic restrictions on …
Noise enhanced stability in magnetic systems
2009
In this paper noise enhanced stability in magnetic systems is studied by both an Ising-type model and a Preisach–Arrhenius model as well as a dynamic Preisach model. It is shown that in one nonequilibrium Ising system noise enhanced stability occurs and that dynamic Preisach model has the capability to predict the occurrence of noise enhanced stability in magnetic systems. On the contrary, in a Preisach–Arrhenius model of a single quadrant magnetic material, noise enhanced stability is not detected.
Improved Neural Networks with Random Weights for Short-Term Load Forecasting.
2015
An effective forecasting model for short-term load plays a significant role in promoting the management efficiency of an electric power system. This paper proposes a new forecasting model based on the improved neural networks with random weights (INNRW). The key is to introduce a weighting technique to the inputs of the model and use a novel neural network to forecast the daily maximum load. Eight factors are selected as the inputs. A mutual information weighting algorithm is then used to allocate different weights to the inputs. The neural networks with random weights and kernels (KNNRW) is applied to approximate the nonlinear function between the selected inputs and the daily maximum load…
A novel four-quadrant power supply for low-energy correction magnets
2003
Abstract This paper describes an efficient power supply to feed low-energy correction magnets in particle accelerator applications, where a controlled current with trapezoidal profile and four-quadrant operation is needed. The selected design is based on an AC–DC matrix converter topology, which uses the Space Vector Modulation (SVM) technique to obtain a near unity power factor at the AC input and output DC current regulation. This topology allows performing high-frequency isolation, while four-quadrant operation is maintained, and reducing volume and weight as compared with the classical thyristor (SCR)-based technology. Control tasks are implemented on an all-digital control card: output…
Focusing ions by viscous drag and weak electric fields in an ion guide
1991
Abstract A new special ion guide has one or more grids at low electric potential in the space between nozzle and skimmer. Viscous drag caused by helium flow is used together with weak electric fields for focusing ions through the skimmer. A typical yield through the skimmer is 75% of that through the nozzle. The focusing device is called a “squeezer”. Most properties of the “squeezer ion guide” are similar to those of an ordinary ion guide. Because the kinetic energies are only of the order of 10 eV, however, problems caused by ion scattering are greatly reduced as compared to ordinary ion guides.
Studying exotic nuclides close to the N = Z line at the HIGISOL facility
2003
The ion guide [1, 2] for heavy-ion fusion-evaporation reactions (HIGISOL) which was developed by Beraud et al. [3] has been implemented at the IGISOL facility in Jyvaskyla [4]. This system was modified over the past 5 years. Figure 1 shows the present set-up. The HIGISOL takes advantage of the different angular distributions of primary beam and reaction products: the primary beam is stopped in front of the stopping chamber and the reaction products enter the stopping chamber through a thin foil passing the beam stop. This so called “shadow” method removes the plasma effect since the primary beam is not ionising the stopping gas. In order to improve ion optical properties, mainly to reduce t…
Enhanced detection techniques of orbital angular momentum states in the classical and quantum regimes
2021
Abstract The orbital angular momentum (OAM) of light has been at the center of several classical and quantum applications for imaging, information processing and communication. However, the complex structure inherent in OAM states makes their detection and classification nontrivial in many circumstances. Most of the current detection schemes are based on models of the OAM states built upon the use of Laguerre–Gauss (LG) modes. However, this may not in general be sufficient to capture full information on the generated states. In this paper, we go beyond the LG assumption, and employ hypergeometric-Gaussian (HyGG) modes as the basis states of a refined model that can be used—in certain scenar…