446494 research outputs found
Sort by
Measuring the dark matter self-interaction cross-section with deep compact clustering for robust machine learning inference
No full textWe have developed a machine learning algorithm capable of detecting ‘out-of-domain data’ for trustworthy cosmological inference. By using data from two separate suites of cosmological simulations, we show that our algorithm is able to determine whether ‘observed’ data is consistent with its training domain, returning confidence estimates as well as accurate parameter estimations. We applied our algorithm to 2D images of galaxy clusters from the BAHAMAS-SIDM and DARKSKIES simulations with the aim of measuring the self-interaction cross-section of dark matter. Through deep compact clustering, we constructed an informative latent space where galaxy clusters were mapped to the latent space forming ‘latent clusters’ for each simulation, with the location of the latent cluster corresponding to the macroscopic parameters, such as the cross-section, σDM/m. We then passed through mock observations, where the location of the observed latent cluster informed us of which properties are shared with the training data. If the observed latent cluster shares no similarities with latent clusters from the known simulations, we can conclude that our simulations do not represent the observations and discard any parameter estimations. This thus provides us with a method of measuring measure machine learning confidence. This method serves as a blueprint for transparent and robust inference that is in demand in scientific machine learning
IoT Based Sign Language Detection and Voice Conversion with Image Processing
No full textThe paper introduces an IoT-enabled system for real-time sign language recognition and voice conversion to improve communication for people with hearing or speech impairments. Using deep learning with TensorFlow, the model accurately detects hand gestures from American and Chinese Sign Language through a standard webcam, with OpenCV handling image processing and pyttsx3 converting recognized signs into speech. An ESP32 microcontroller transmits the interpreted data over Wi-Fi and hosts a mobile-friendly web page, eliminating the need for extra hardware or dedicated apps. This low-cost, efficient solution achieves high real-time accuracy, offering both audio and visual feedback, and showcases the effective integration of AI and IoT in bridging communication gaps
Parametric Optimization and Performance Analysis of a Central Receiver Solar Power Plant: A Numerical Study for Enhanced Solar Field and Receiver
No full textIn the context of the energy transition and the search for sustainable solutions, central receiver solar power plants have become a promising technology for generating green electricity. The present case study aims to identify the optimal dimensional parameters of a solar tower plant located in southern Morocco. Typical Meteorological Year (TMY) data were collected for the targeted region to ensure a realistic representation of local climatic conditions. The modelling of the optical behavior of the heliostat field and the receiver is carried out using a simulation approach based on the Monte Carlo Ray Tracing (MCRT) method. The main objective is to identify an optimal configuration by varying several key parameters, such as the layout of the heliostats, the tower height, and the receiver dimensions. The results obtained show that an increase in the tower height leads to a growth in the number of accessible heliostats and therefore an expansion of the mirror field. Optimal performance is achieved for a tower height between 180 and 200 meters. For this height of the tower the optical losses due to dissipation effects and atmospheric attenuation are at their maximum, reaching approximately 5.05% and 8.9%, respectively. On the other hand, losses due to blocking/shading effects and the cosine effect reach their minimum values, at approximately 1% and 20%, respectively
Fission fragment distributions of the excited compound nucleus
No full textThe present study focuses on the measurement and analysis of the cross sections of 35 fission like residues (with mass numbers 74 A 111) and 10 fusion-like residues in the 14N+175Lu reactions at energies 75.8 MeV, 79.7 MeV and 84.0 MeV, respectively. The measurements have been carried out using the recoil catcher technique followed by off-beam activation -ray spectroscopy. Analysis of fusion reaction data reveals the presence of complete and incomplete fusion mechanisms in the 14N+175Lu reactions at these energies. In addition to the fusion contribution, a substantial amount of fission following the complete and incomplete fusion is also observed. The experimentally deduced post-fission observables mass () and charge () dispersion parameters from the isotopic and isobaric distributions of fission fragments are found to be in good agreement with their values reported in the literature and calculated theoretically as well. Such consistency refers to the production of these fission fragments from the equilibrated compound nucleus () in 14N+175Lu reactions. This finding is also verified by the observations of the symmetric mass distributions of fission fragments which are found to have a single-humped Gaussian shape and their mass variances () follow the exponential trend with excitation energy, a characteristic of compound nucleus fission. The consistency in the isotopic, isobaric and mass distributions of the fission fragments in the 14N+175Lu system enhances an additional understanding of compound nucleus fission in heavy-ion reactions
New measurement of
No full textCr and Cr are very relevant in criticality safety benchmarks related to nuclear reactors. The discrepancies of up to 30% between the neutron capture cross section evaluations have an important effect on the and in criticality benchmarks particularly sensitive to chromium. In this work, the Cr(n,) cross sections are to be determined between 1 and 100 keV with an 8–10% accuracy following the requirements of the NEA High Priority Request List (HPRL) to solve the current discrepancies. We have measured these reactions by the time-of-flight technique at the EAR1 experimental area of the n_TOF facility, using an array of four CD detectors with very low neutron sensitivity. The highly-enriched samples used are significantly thinner than in previous measurements, thus minimizing the multiple-scattering effects. We have produced, and analysed with the R-matrix analysis code SAMMY, capture yields featuring 33 resonances of Cr and 51 of Cr with an accuracy between 5% and 9%, hence fulfilling the requirements made by the NEA. The differential and integral cross sections have been compared to previous data and evaluations. The new measured Cr(n,) cross sections provide a valuable input for upcoming evaluations, which are deemed necessary given that the results presented herein do not support the increase in both cross sections proposed in the recent INDEN evaluation
–
No full textWe study the spin-averaged elastic scattering cross sections of the – system using a QCD-motivated potential framework and the Resonating Group Method (RGM). The interaction incorporates a geometric form factor based on the minimal-area variable , supported by lattice studies of multi-quark Wilson-loop geometries. Compared with the conventional Gaussian form, the –based overlap yields slightly larger cross sections near threshold, reflecting its harder short-distance structure and its closer compatibility with the compact wavefunction. A sensitivity analysis over the physically motivated range –0.70 shows that while the magnitude varies smoothly with the stiffness parameter, the rapid suppression with increasing kinetic energy remains stable across all values. The results confirm that short-range dynamics dominate bottomonium–bottomonium interactions, consistent with expectations from generalized Gauss-law potentials and leading-twist QCD analyses. The modest near-threshold enhancement may offer useful input for interpreting suppression patterns in heavy-ion collisions, where quarkonium interactions with the medium play an important role
A method to measure the angular differential cross section of (
No full textA method to measure the angular differential cross section of (n, ) reactions using gridded ionization chamber (GIC) and gas sample was established. The anode signal rise time is used to get the projection length of the track in the direction of the electric field (related to the emission angle). The anode signal amplitude is used to distinguish the events with the same anode signal rise time but different emission angles because of the PHD (pulse height defect) effect. Using this method, the angular differential cross sections of the O(n, )C reaction at 10.45 MeV were measured based on the HI-13 tandem accelerator of China Institute of Atomic Energy (CIAE). The GIC was used as the charged particle detector, with working gas of 3.0 atm Kr+4.0%CO. The oxygen atoms in the CO were used as the sample. Anode signal amplitude vs the anode signal rise time two-dimensional spectrum was used in this method to determine the count of the O(n, )C reaction events. A UO sample inside the GIC was used to determine the neutron fluences and an EJ-309 scintillation detector was placed on the beam line to measure the neutron energy spectra to correct the events induced by the low-energy neutrons