-
Towards Reflectivity profile inversion through Artificial Neural Networks
Authors:
Juan Manuel Carmona-Loaiza,
Zamaan Raza
Abstract:
The goal of Specular Neutron and X-ray Reflectometry is to infer materials Scattering Length Density (SLD) profiles from experimental reflectivity curves. This paper focuses on investigating an original approach to the ill-posed non-invertible problem which involves the use of Artificial Neural Networks (ANN). In particular, the numerical experiments described here deal with large data sets of sim…
▽ More
The goal of Specular Neutron and X-ray Reflectometry is to infer materials Scattering Length Density (SLD) profiles from experimental reflectivity curves. This paper focuses on investigating an original approach to the ill-posed non-invertible problem which involves the use of Artificial Neural Networks (ANN). In particular, the numerical experiments described here deal with large data sets of simulated reflectivity curves and SLD profiles, and aim to assess the applicability of Data Science and Machine Learning technology to the analysis of data generated at neutron scattering large scale facilities. It is demonstrated that, under certain circumstances, properly trained Deep Neural Networks are capable of correctly recovering plausible SLD profiles when presented with never-seen-before simulated reflectivity curves. When the necessary conditions are met, a proper implementation of the described approach would offer two main advantages over traditional fitting methods when dealing with real experiments, namely, 1. sample physical models are described under a new paradigm: detailed layer-by-layer descriptions (SLDs, thicknesses, roughnesses) are replaced by parameter free curves $ρ(z)$, allowing a-priori assumptions to be fed in terms of the sample family to which a given sample belongs (e.g. "thin film", "lamellar structure", etc.) 2. the time-to-solution is shrunk by orders of magnitude, enabling faster batch analyses for large datasets.
△ Less
Submitted 24 January, 2021; v1 submitted 15 October, 2020;
originally announced October 2020.
-
Overlapping inflows as catalysts of AGN activity - II: Relative importance of turbulence and inflow-disc interaction
Authors:
Juan Manuel Carmona-Loaiza,
Monica Colpi,
Massimo Dotti,
Riccardo Valdarnini
Abstract:
The main challenge for understanding the fuelling of supermassive black holes in active galactic nuclei is not to account for the source of fuel, but rather to explain its delivery from the boundaries of the black hole sphere of influence (10-100 pc) down to sub-parsec scales. In this work, we report on a series of numerical experiments aimed at exploring in further depth our model of "overlapping…
▽ More
The main challenge for understanding the fuelling of supermassive black holes in active galactic nuclei is not to account for the source of fuel, but rather to explain its delivery from the boundaries of the black hole sphere of influence (10-100 pc) down to sub-parsec scales. In this work, we report on a series of numerical experiments aimed at exploring in further depth our model of "overlapping inflow events" as catalysts for rapid accretion, seeding a turbulent field in the infalling gas. We initially set a gaseous shell in non-equilibrium rotation around a supermassive black hole. After infall, the shell stalls in a disc-like structure. A second shell is then set in either co-rotation or counter-rotation with respect to the first and is let to impinge on the previously-formed disc. We find that combined turbulence and overlap significantly enhance accretion in counter-rotating inflows, while turbulence dominates for co-rotating inflows. The leftovers of overlapping inflows are warped nuclear discs, whose morphology depends on the relative orientation and angular momentum of the disc and the shell. Overlapping inflows leave observational signatures in the gas rotation curves.
△ Less
Submitted 31 July, 2015;
originally announced August 2015.
-
Overlapping Inflow Events as Catalysts for Supermassive Black Hole Growth
Authors:
Juan Manuel Carmona-Loaiza,
Monica Colpi,
Massimo Dotti,
Riccardo Valdarnini
Abstract:
One of the greatest issues in modelling black hole fuelling is our lack of understanding of the processes by which gas loses angular momentum and falls from galactic scales down to the nuclear region where an accretion disc forms, subsequently guiding the inflow of gas down to the black hole horizon. It is feared that gas at larger scales might still retain enough angular momentum and settle into…
▽ More
One of the greatest issues in modelling black hole fuelling is our lack of understanding of the processes by which gas loses angular momentum and falls from galactic scales down to the nuclear region where an accretion disc forms, subsequently guiding the inflow of gas down to the black hole horizon. It is feared that gas at larger scales might still retain enough angular momentum and settle into a larger scale disc with very low or no inflow to form or replenish the inner accretion disc (on ~0.01 pc scales). In this paper we report on hydrodynamical simulations of rotating infalling gas shells impacting at different angles onto a pre-existing, primitive large scale (~10 pc) disc around a super-massive black hole. The aim is to explore how the interaction between the shell and the disc redistributes the angular momentum on scales close to the black hole's sphere of influence. Angular momentum redistribution via hydrodynamical shocks leads to inflows of gas across the inner boundary, enhancing the inflow rate by more than 2-3 orders of magnitude. In all cases, the gas inflow rate across the inner parsec is higher than in the absence of the interaction, and the orientation of the angular momentum of the flow in the region changes with time due to gas mixing. Warped discs or nested misaligned rings form depending on the angular momentum content of the infalling shell relative to the disc. In the cases in which the shell falls in near counter-rotation, part of the resulting flows settle into an inner dense disc which becomes more susceptible to mass transfer.
△ Less
Submitted 22 December, 2013; v1 submitted 9 December, 2013;
originally announced December 2013.
